Zoo Biology 32: 291–298 (2013) RESEARCH ARTICLE Variation in the Composition of Milk of Asian Elephants (Elephas maximus) Throughout Lactation Frances N. Abbondanza,1* Michael L. Power,2,3 Melissa A. Dickson,4 Janine Brown,1 and Olav T. Oftedal5 1 Department of Center for Species Survival, Smithsonian Conservation Biology Institute, Front Royal, Virginia Nutrition Laboratory, Conservation Ecology Center, Smithsonian Conservation Biology Institute, National Zoological Park, Washington, District of Columbia 3 Research Department, American College of Obstetricians and Gynecologists, Washington, District of Columbia 4 Dickerson Park Zoo, Springfield, Missouri 5 Smithsonian Environmental Research Center, Edgewater, Maryland 2 We investigated milk nutrient composition from three Asian elephant cows over the first 3 years of lactation, including two consecutive lactations in one cow. Body mass gain is presented for three calves during the first year. Milk samples (n = 74) were analyzed for dry matter (DM), fat, crude protein (CP), sugar, ash, calcium (Ca), phosphorus (P), and potassium (K); gross energy (GE) was calculated. Concentrations of most nutrients changed over lactation: DM, fat, CP, Ca, P, and GE were positively correlated to calf age; sugar was negatively correlated to calf age. GE doubled between birth (1 kcal/g) and 2 years of age (2 kcal/g). After accounting for calf age, GE, fat, Ca, and P concentrations differed among the cows. Milk composition also differed between two lactations from the same cow. When milk nutrients were expressed on a mg per kcal basis, the pattern changes: CP, Ca, and P remained relatively constant over lactation on a per energy basis. Calf mass quadrupled over the first year of life; mass gain was linear at 0.9 kg/day. Asian elephant milk composition is variable, both across lactations and between cows, complicating efforts to determine representative values for comparative studies and for the formulation of elephant milk formulas. The fact that CP, Ca, and P were all relatively constant when expressed on a per energy basis may be of biological significance. The increase in nutrient density over lactation undoubtedly limits maternal water loss, reducing the volume of milk necessary to support the calf. Zoo Biol. 32:291–298, 2013. © 2012 Wiley Periodicals, Inc. Keywords: lactation; milk constituents; milk replacers INTRODUCTION Lactation has evolved as an efficient means to transfer nutrients, enzymes, growth factors, and immune constituents from mother to young [Jensen, 1995]. The diversity of mammalian radiations is reflected in the diversity of evolved lactation strategies. Although the major constituents of milk (lipid globules, α-, β-, and κ- caseins, major whey proteins, lactose and/or oligosaccharides, minerals, vitamins, and water) are generally similar across species [Oftedal, 2012], a comparison of milk composition among mammals reveals large differences in the relative amounts of these constituents [Jenness and Sloan, 1970; Oftedal, 1984; Oftedal and Iverson, 1995; Urashima et al., 2001]. Some studies have found significant variation in milk composition among individuals, and even among lactations from the same individual [e.g., Power et al., 2002]. In most species, milk composition © 2012 Wiley Periodicals, Inc. changes over the course of lactation [Oftedal, 1984; Oftedal and Iverson, 1995]; in some species, milk composition changes over the course of an individual nursing bout as well [Oftedal, 1984]. Thus, characterizing the milk composition of a species can be a complex process. Elephants, the largest terrestrial mammals, make longterm maternal investments during their extended lives. The ∗ Correspondence to: Frances Abbondanza, Endocrine Technician, Smithsonian Conservation Biology Institute, 1500 Remount Road, Front Royal, VA 22630. E-mail: Parkern@si.edu Received 18 December 2011; Revised 18 March 2012; Accepted 23 March 2012 DOI 10.1002/zoo.21022 Published online 15 May 2012 in Wiley Online Library (wileyonlinelibrary. com). 292 Abbondanza et al. gestation of Asian (Elephas maximus) and African (Loxodonta africana) elephants lasts approximately 660 days, and females give birth to calves weighing approximately 75–115 kg [Eltringham, 1997; Laws and Parker, 1968; Poole, 1997; Shoshani, 1992]. It is essential for calves to nurse immediately following birth to establish a bond between mother and calf, and to receive colostral substances that are supplied within the first few days. Elephant calves are solely dependent on receiving nourishment from their mother’s milk for the first 3–6 months [Lee and Moss, 1986]. Lactation lasts anywhere from 2 to 8 years and weaning is usually contingent upon when the cow gives birth again [Ochs et al., 2001; Sheldrick, 1990; Welsch, 1998]. Even if milk is not necessary for survival beyond 2 years, extended maternal care may be important for maintaining growth rates, body condition, and ultimate reproductive ability of offspring [Sukumar, 2003]. It is intriguing that elephants share multiple similarities with primates and more specifically, humans, including the complete dependency of suckling young on their mothers for the first few years of life, a slow-growth rate at similar developmental stages, and a long-life span [McCullagh and Widdowson, 1970; Poole, 1997]. Mammary glands undergo lactogenesis at parturition, as mammary genes associated with milk production are upregulated under endocrine influence [Lemay et al.,]. Elephant mammary glands are lobular structures that produce a complex secretion by a variety of secretory mechanisms, including exocytosis of the aqueous phase of milk (including most proteins) and an apocrine-like pinching off of membrane-bound milk fat globules (MFG) [Welsh, 1998]. The proximate composition of elephant milk was first studied in the 19th century [Doremus, 1881]. A number of constituents have been examined in African and/or Asian elephant milks, including casein micelles, casein and whey protein fractions, amino acid patterns, MFG size distribution, fatty acid patterns, organic acids, mono-, di-, and oligosaccharides, macrominerals, and some fat- and water-soluble vitamins [Kunz et al., 1999; Mainka et al., 1994; McCullagh and Widdowson, 1970; Osthoff et al., 2005; Peters et al., 1972; Simon, 1959; Uemura et al., 2006; Welsh, 1998]. However, little is known about how elephant milk changes over the course of lactation; available longitudinal studies over the course of lactation are both limited in scope and inconsistent [Mainka et al., 1994; McCullagh and Widdowson, 1970; Peters et al., 1972; Simon, 1959]. We undertook the first study of longitudinal change in milk composition across multiple individuals of the Asian elephant, E. maximus. The lactation strategy, milk production, and milk composition of the Asian elephant are fundamental components of the reproductive and nutritional biology of this species, and knowledge gained may benefit the captive born elephant population by providing information for the development of appropriate hand-rearing protocols for orphaned or abandoned calves. This paper presents data on the composition of Asian elephant milk collected from three Asian elephants at the Dickerson Park Zoo (Springfield, MO) over four lactation periods. MATERIALS AND METHODS Animals and Sample Collection Milk composition was studied during four lactations of three Asian elephants at the Dickerson Park Zoo, Springfield, Mo (Table 1). Two elephant cows (Connie, Patience) were studied in one lactation, and one (Moola) in two sequential lactations. The lactating elephants were fed a daily diet consisting of a concentrate feed (HMS elephant supplement 9072, Higginbottom Management Services, Bluffton IN 46714) plus good quality grass hay (orchard grass, brome, timothy or a mixture; average 8–12% crude protein) and small amounts (0–2 kg) of produce, e.g., as training rewards. The concentrate was consumed at the offered level of about 4–5 kg per day by elephants Moola and Patience; Connie was less predictable in intake and thus at times was offered more concentrate (up to about 8 kg) but on average also consumed about 4–5 kg per day. Hay consumption was consistently about 45–55 kg per elephant per day in all elephants. The pellets (declared values: 24% crude protein, 4.9% fat, 12% acid detergent fiber [min], 1.0% calcium, 1.0% phosphorus, 0.7 ppm selenium, 15,000 IU per kg vitamin A, 3,600 IU per kg vitamin D3, 900 IU per kg vitamin E) were formulated to be fed at 10% of diet dry matter; the declared nutrient levels met or exceeded the recommended concentrations for elephant concentrate pellets (Ullrey et al. 1997) for all nutrients except calcium (83% of recommended), manganese (94%), selenium (88%), panthothenate (93%), and choline (86%); copper was also marginally high (116% of recommended maximum level). Given that good quality hay was provided, the entire diet was considered nutritionally complete [Oftedal et al., 1996; Ullrey et al., 1997]. The elephant cows were conditioned to permit milk collection. Calves were encouraged to latch onto the breast and nurse for 2–5 min in order to stimulate milk let down and then TABLE 1. Summary of milk samples collected from elephants at the Dickerson Park Zoo, Springfield, MO Dam Connie Patience Moola Moola a YOB Calf DOBa Sex of calf No. of samples ~1964 ~1974 1981 1981 02/02/1995 05/17/1998 07/02/1996 11/28/1999 F M F M 16 14 22 22 All calves were sired by Onyx, Studbook #104, now deceased. Zoo Biology Day of 1st milk sample 609 3 93 9 Day of last sample 1136 321 710 654 Asian Elephant Milk Composition 293 briefly displaced to allow manual milk collection. Milk samples (2–30 ml) were collected monthly or bimonthly between 1996 and 2001. The dates of parturition, number of samples collected, and length of sample collection for each lactation period are presented in Table 1. Note that for the first cow studied, milk collection was not started until about 20 months post-partum, but in subsequent lactations milk sampling began at 3–93 days post-partum. Upon collection, milk was placed in air-tight plastic scintillation vials and immediately frozen at –20°C; samples were kept frozen until analyzed. non-protein nitrogen [Oftedal, 1984]; however, GE values calculated by this formula were not different from GE values determined by bomb calorimetry for rhesus macaque and Weddell seal milk samples [Hinde et al., 2009, Eisert and Oftedal, unpublished data]. The GE values were used to calculate the inverse parameter: grams of milk to equal 1 kcal of milk (1/GE). The amount of any milk constituent (expressed in %) multiplied by 10 × (1/GE) equals the mg of that constituent associated with 1 kcal. Minerals Sample Analyses Milk samples were analyzed at the Nutrition Laboratory at the Smithsonian National Zoological Park (NZP), Washington, DC using standard methods [Oftedal and Iverson, 1995]. All milk constituents were analyzed in duplicate and results are reported on a fresh weight basis. Milk constituents are expressed on a gram-per-gram basis (in %) and as the mg of nutrient per kcal, based on the calculated GE values. Proximate analyses For dry matter determination, milk samples were aliquoted, weighed (mean 18.98 mg; range 16.96–24.51 mg), and dried in a forced air-drying oven for 3 hr at 100°C, and then reweighed. Total nitrogen (TN) was determined for these dried samples (mean 4.69 mg; range 1.30–7.60 mg) using a carbon, hydrogen, and nitrogen (CHN) elemental gas analyzer (Model 2400, Perkin Elmer, Norwalk, CT) at a combustion temperature of 950°C with supplemental oxygen boosts of 2 sec to ensure complete combustion. This method has been validated against the macro Kjeldahl procedure with nitrogen recovery around 98–99%, and has been used at NZP to measure milk nitrogen for a wide variety of species. The TN value was multiplied by 6.38 to determine the amount of crude protein (CP) found in the milk [Jones, 1931]. Total lipid was measured on approximately 0.15 g milk (mean 138.7 mg; range 126.2–151.9 mg) using a micro modification of the Roese–Gottlieb procedure involving sequential extractions with diethyl ether and petroleum ether after disruption of milk fat globules with ammonium hydroxide and ethanol. Total sugar or carbohydrate were analyzed in samples (mean 13.8 mg; range 10.4–35.0 mg) diluted into approximately 25 g distilled water by the phenol–sulphuric acid colormetric procedure [Dubois et al., 1956; Marier and Boulet, 1959] using lactose monohydrate to prepare standards; absorbance was read at 490 nm with a UV-visible spectrophotometer (Beckman DU Model 640, Beckman Coulter, Fullerton, CA) equipped with an automatic sipper tube. Gross energy Gross energy (GE) was calculated assuming 9.11 kcal/g for fat, 5.86 kcal/g for protein, and 3.95 kcal/g for sugar, and is expressed as kcal/g milk. This calculation will slightly overestimate GE because it does not account for For total mineral estimates (ash) 1 g milk samples (range 0.980–1.024 g) were dried in crucibles and then combusted in a muffle furnace at 500°C for 4 hr. For specific mineral analyses, liquid milk samples (mean 1.98 g; range 1.02–2.21 g) were digested using nitric and perchloric acid on a hot plate contained within a perchloric-acid rated fume hood. Mineral digests were diluted in distilled, deionized water. Diluted digests were analyzed for calcium (Ca) and potassium (K) by flame atomic absorption spectrophotometry at 422.7 nm and 766.5 nm wavelengths, respectively, using a Model 800 Perkin Elmer Analyst Flame/Furnace Atomic Absorption Spectrophotometer (Perkin Elmer Co, Waltham, MA). Lanthanum chloride (1,000 ppm) and potassium chloride (1,000 ppm) were used as modifiers for the Ca assay; cesium chloride (1,000 ppm) was used as a modifier for the K assay. Phosphorus (P) was analyzed by the AOAC–Modified Gomori method using a UV-visible spectrophotometer (Beckman DU Model 640, Beckman Coulter, Fullerton, CA) with absorbance measured at 450 nm. Calf Weights We retrospectively examined management records in order to calculate the rates of mass gain of three calves, two of which were female and one male. Statistical Analysis Summary values are expressed as mean ± SEM. Pearson correlation coefficients were calculated to examine the associations between calf age and milk constituents and among milk constituents. Multivariate analysis of covariance with calf age as the covariate was used to test for differences in milk composition between cows. Because one cow was represented across two different lactations the MANOVA was run three times: with cow as the categorical variable (n = 3), with lactation as the categorical variable (n = 4), and using only the individual Moola’s samples to test for differences between her two lactations. Variation in the milk constituents was examined both on a gram-per-gram basis and when expressed as mg per kcal. Rate of calf mass gain over the first year of life was estimated using linear regression. Variation in growth among the three calves was examined using ANCOVA. Statistical significance was set at P < 0.05. Zoo Biology 294 Abbondanza et al. RESULTS All nutrients analyzed except for K showed significant associations with calf age (Table 2). Strong positive correlations were shown between the age of calf and the following milk constituents: fat, CP, ash, Ca, and P (P < 0.01). Water (=100 dry matter) and sugar contents were negatively correlated with calf age, and with the other milk constituents (P < 0.01). The GE of milk increased from about 1 kcal/g in the first months of lactation to approximately 2 kcal/g in the second year of lactation, but then appeared to plateau. The best-fit curve for the data was a power function (given in Figure legend) and the slope of the linear regression for samples after calf age of 500 days was not significantly different from zero (Fig. 1). The mean percent of fat and CP in milk increased over time while the percent sugar decreased (Fig. 2). The percent of Ca and P in milk increased with calf age (Fig. 3); K did not change (data not shown). Fat, sugar, CP, and GE varied with calf age (P < 0.001). Among cows, GE (P < 0.001) and fat (P = 0.001) were significantly different; but sugar (P = 0.125), and CP (P = 0.132) were not. For the minerals, Ca (P < 0.001) and P (P = 0.026) varied significantly among cows, and with calf age (P < 0.001 for both); but K did not (P = 0.114 and 0.099, respectively). Comparing Moola’s two lactations, calf age was a significant factor for all nutrients including K (P < 0.005 for all). CP (P = 0.003), P (P = 0.001), and K (P < 0.001) were significantly different between lactations; GE (P = 0.121), fat (P = 0.169), sugar (P = 0.058), and Ca (P = 0.090) were not. The overall measure of significance for the MANCOVA was P < 0.001, indicating a significant difference in milk composition between these two lactations in the same cow. Accordingly, we ran the MANCOVA for the three elephants treating each of Moola’s lactations as independent events (four lactations). Calf age was a significant factor for all milk constituents (P < 0.001) except K (P = 0.237). All TABLE 2. Correlation coefficients for milk constituents and calf age Calf age GE GE Water Fat CP Sugar Ash Ca p 0.730 −0.718 −0.963 Fat 0.728 0.997 −0.959 CP 0.774 0.768 −0.750 0.730 Water −0.856 −0.767 0.751 −0.772 −0.846 Ash 0.789 0.629 −0.612 0.624 0.762 −0.787 Ca 0.784 0.605 −0.565 0.613 0.633 −0.773 0.819 P 0.841 0.842 −0.84 0.839 0.807 −0.863 0.837 Sugar 0.781 0.190 0.220 −0.230 0.225 0.198 −0.254 0.504 0.449 0.323 P = 0.105 P = 0.059 P = 0.049 P = 0.054 P = 0.090 P = 0.029 P < 0.001 P < 0.001 P = 0.005 K n = 74 except for Ash where n = 45; P < 0 .001 for all correlations except for K, where P-values are provided. Fig. 1. The gross energy (GE) of milk expressed as kcal/g increased with calf age. Trend line is from the equation: GE = 0.512 × (calf age)0.194, R2 = 0.557. Zoo Biology Fig. 2. Mean and SEM for fat, protein, and sugar in Asian elephant milk by calf age. Asian Elephant Milk Composition 295 constituents varied significantly among lactations (GE, fat, Ca, P, and K, P = 0.001; CP, P = 0.007; sugar, P = 0.024). When milk constituents were expressed as mg per kcal, the pattern changes somewhat. Fat increased with calf age (r = 0.693, P < 0.001) while water (r = −0.693, P < 0.001), and sugar decreased (r = –0.826, P < 0.001); however, the mg of protein per kcal remained relatively constant (r = –0.120, P = 0.310). Milk fat increased from an average of about 65 mg/kcal in the first months of lactation to 85 mg/kcal when the calf was 3 years old, and milk sugar decreased from 55 mg/kcal to less than 15 mg/kcal; milk CP was relatively constant over three years of lactation at about 30 mg/kcal (Fig. 4). Among the minerals, ash (r = –0.084, P = 0.583) and Ca (r = 0.120, P = 0.307) were constant at about 4.2 mg/kcal and 0.9 mg/kcal, respectively. Phosphorus paralleled Ca at about 0.5 mg/kcal, although the slight rise in P over lactation was significant (r = 0.234, P = 0.045, Fig. 5). Consistent with the lack of an absolute change in K over lactation, its value expressed as mg/kcal declined with calf age (r = –0.538, P < 0.001). Water per kcal decreased exponentially from about 900 mg/kcal in the first months of lactation to about 350 mg/kcal after 2 years (Fig. 6). When the milk constituents were expressed as mg per kcal in the MANCOVAs [whether treating the two lactations of Moola as independent events (n = 4) or not (n = 3)], calf age was not a significant factor for CP or ash (P > 0.3 in all cases). Calf age was a significant factor for the other nutrients (P < 0.05), except for Ca, where it was only a trend (P = 0.059 and P = 0.056, respectively). The different lactations were significantly different for all nutrients (P < 0.001) except protein per energy (P = 0.076) and P (P = 0.393). When the two Moola lactations were combined, cow was a significant factor for mg protein per kcal (P = 0.032) and the other nutrients (P < 0.001), except for P (P = 0.231). Mass gain of the calves in the first year of life was remarkably similar and linear (Fig. 7). The calves, on average, gained 0.9 kg/day. There was no significant difference between calves (P = 0.613). Calf mass roughly quadrupled Fig. 3. Mean and SEM for calcium and phosphorus in Asian elephant milk by calf age. Fig. 4. Fat, protein, and sugar expressed on a per energy basis by calf age. Fat increased and sugar decreased. Protein remained constant. Fig. 5. Calcium and phosphorus expressed on a per energy basis by calf age. Fig. 6. The mg of water per kcal of milk exponentially declines with calf age. Zoo Biology 296 Abbondanza et al. Fig. 7. Growth over first year of life for the three calves with corresponding milk composition data; cow’s name in parentheses. in the first year of life. Based on feeding observations calves begin eating solid food by 4 months and by 7 months food intake is a significant amount [Dickson, 2000, personal communication]. Based on these observations we defined peak lactation to be 4–7 months, the time period where calves begin to change over from milk to solid food as a major portion of their diet. Mean values for milk constituents at peak lactation were (mean ± SEM; n = 9): DM = 19.6 ± 1.2%, fat = 8.3 ± 1.0%, CP = 3.6 ± 0.3%, sugar = 4.9 ± 0.3%, and GE = 1.2 ± 0.1 kcal/g. However, calves continue to nurse for years. DISCUSSION The results from this study indicate that Asian elephant milk is variable, both between cows and as the calf aged. Milk composition even varied between lactations within the same cow. As the volumes of milk samples collected in this study (2–30 ml) were relatively small compared to the substantial but unknown amounts of milk consumed by the calf during each nursing bout [Andrews et al., 2005], our samples did not represent entire mammary contents. In humans, horses, and some ruminants, milk fat content rises during the course of mammary evacuation, such that the first milk drawn is low in fat and the residual milk after nursing is high in fat, with a range of about 3-fold [Oftedal, 1984]. All of our samples were taken by interrupting nursing bouts and are thus unlikely to encompass either extreme (initial vs. final), but variation in the time at which the sample was collected during the nursing bout could have contributed to the variability in fat observed in our samples. Nevertheless, there was a general trend for fat and CP percentages to increase, and water and sugar percentages to decrease, over the first year of calf life (Table 2 and Fig 2). This is consistent with trends previously observed in milk (n = 30) obtained from wild African elephants from 10 to 36 months postpartum [McCullagh and Widdowson, 1970], and in milk (n = 7) obtained from working Asian elephants Zoo Biology from 2 to 18 months [Simon, 1959], but not in milk (n = 10) collected from zoo Asian elephants from 0.5 to 9 months [Mainka et al., 1994]. Other reports on small numbers of elephant milk samples [e.g., Doremus, 1881; Osthoff et al., 2005; Peters et al., 1972] include considerable compositional variation but no clear trends over lactation stage, perhaps because sample numbers were too low. Given the large variability in the composition of elephant milk samples, large numbers of samples are needed to accurately document composition. Some caution is also necessary in comparing our findings with previous papers due to differences in analytical methodology. For example, the milks of both Asian and African elephants contain a substantial proportion of oligosaccharides as well as the characteristic milk sugar, lactose [Kunz et al., 1999; Osthoff et al., 2005; Uemura et al., 2006]. The phenol-sulfuric acid method used herein measures total sugars and thus incorporates oligosaccharides, although it may be affected to some extent by the proportional composition of different hexose, hexosamine, fucose, and sialic acid units present in the oligosaccharides [Oftedal and Iverson, 1995]. However, reducing sugar methods as used by Simon [1959], and the anthrone method as used by McCullagh and Widdowson [1970] underestimate total sugar if oligosaccharides are present but not hydrolyzed prior to analysis [Oftedal and Iverson, 1995]. On the other hand, calculation of sugar by difference as done by Peters et al. [1972] is inherently imprecise and will overestimate sugar if all other constituents are not accounted for. Fat and sugar represent alternative substrates for energy metabolism. Fat has more than twice the energy-pergram compared to sugar such that an increase in milk fat results in a substantial rise in milk GE. Sugar synthesis, on the other hand, results in osmotic movement of water into the Golgi apparatus, leading to increased volume of secretory vesicles and an increase in milk volume, diluting nutrient density [Shennan and Peaker, 2000]; this osmotic effect may be somewhat compensated if the molecular size of secreted sugars increases (e.g., from lactose to larger oligosaccharides). This is one reason why eutherian mammals with high milk sugar (e.g., equids, rhinos, and most primates) also have dilute milks. Asian and African elephants also have relatively high milk-oligosaccharide levels, as do primates [Goto et al., 2010; Kunz et al., 1999; Osthoff et al., 2005; Uemura et al., 2006; Urashima et al., 2009]. An increase in fat content coupled to a decrease in sugar content, as occurs over lactation in Asian elephant milk, is an effective mechanism to increase the nutrient density of the milk. The increase in nutrient density over lactation also results in less water transfer from mother to calf for the equivalent energy transfer, and thus limits maternal water loss by restricting the absolute volume of milk necessary to support the nutritional needs of the calf. The amounts of CP, ash, and the minerals Ca and P were relatively constant over lactation when expressed on a per energy basis. Thus, variation in Asian elephant milk composition depends in part upon the units selected. Energetics Asian Elephant Milk Composition 297 has significant explanatory power for many biological phenomena [Blaxter, 1989], and has long been recognized as an appropriate basis for comparison of nutrient levels in milk [Bernhart, 1961; Oftedal, 1984, 1986; Powers, 1933]. If elephant calves nurse to satisfy energy needs, as opposed to a volume limit, nutrient intakes will be determined by milk energy density. Protein, Ca, and P are essential for tissue and bone growth and by being correlated to milk energy their intakes are regulated. It is intriguing that CP, Ca, and P appear to maintain a constant density relative to milk GE. This phenomenon has also been documented in several primate species. Even though there can be considerable variation in the absolute concentration of milk protein among samples from common marmosets, squirrel monkeys, and rhesus macaques, the mg protein per kcal is remarkably constant [Hinde et al., 2009; Milligan et al., 2008; Power et al., 2002; 2008]. Given that daily mass gain of elephant calves is constant in the first year, one would expect daily protein requirements (g per day) for tissue growth also to be relatively constant over this time period [Oftedal, 1986]. However, the quadrupling of calf body mass (and increased calf activity) implies an ever increasing daily energy requirement, most of which must be supplied by milk, at least in the first 7 months of life. Thus one would predict that protein requirements would fall relative to energy as the calf ages, and cows should not need to provide as much protein per kcal later in lactation. It may be that milk serves more as a protein supplement (to compensate for lower protein quality in elephant forage) in later lactation, but this does not explain why the mg protein per kcal would remain constant. Information generated from this study contributes to the limited knowledge of Asian elephant milk, and may be valuable for developing suitable milk replacers for calves that lose or are rejected by their mothers. Milk replacers are often formulated to human or cow’s milk specifications, using a variety of cow’s milk and non-milk constituents. At least one milk-replacer formula has been developed specifically for Asian and African elephant calves, but a variety of other formulas, including human infant formulas, have been used in North America, Europe, the Middle East, and Asia [http://www.elephanttag.org/Professional/professional_nutrition.html]. It would be useful to compare growth rates, digestive performance, and animal health criteria for elephants raised per different hand-rearing protocols and in relation to mother-nursed calves, but published data on the success rates of individual milk replacers are lacking. Our data indicate that a constant milk-replacer formula may not be the most appropriate solution, as the nutrient density of Asian elephant milk increases with calf age. However, the consistency of the mg of CP, Ca, and P per kcal suggests that these parameters may provide useful targets for formulation of milk replacers. Both Asian and African elephant milks also have unusual fatty acid composition, being very rich in medium-chain saturated fatty acids: one-half to two-thirds of all fatty acids in elephant milks are capric (10:0) and lauric (12:0) acids that are presumably of mammary (rather than dietary) origin [McCullagh and Widdowson, 1970; Oftedal and Iverson, 1995; Osthoff et al., 2005; Peters et al., 1972]. They are also rich in oligosaccharides, which may comprise up to half of the total sugar [Kunz et al., 1999; Osthoff et al., 2005]. The difficulty in matching these constituents in cow’s milk-based formulas and the potential growth and health consequences of not doing so have not been evaluated. It is also important to recognize that the initial secretion, colostrum, may supply immunoglobulins, growth factors, and prebiotics (such as oligosaccharides) that are essential to development of normal gastrointestinal function and a healthy immune system, based on what is known in other mammals. Thus, if an elephant cow does not readily accept the calf within the first 24 hr, commercially available immunoglobulin supplements or bovine colostrum may need to be provided to the calf. Maternal plasma can also be collected and administered to the calf, assuring delivery of immunoglobulins within the first few days after birth [Dickson, personal communication, 2006]. Another option, if herd dynamics and reproductive patterns permit, may be to attempt to nurse orphaned or abandoned calves on other females, as has been done in captive bottlenose dolphins in whom lactation can be induced [Ridgway et al. 1995]. Wild elephant calves are known to nurse from not only their own mothers but aunts as well [McKay, 1973; Welsh, 1998], and in captive situations that are usually comprised of artificial herds, calves will occasionally nurse from unrelated females who seem to tolerate this behavior [Dickson, personal observation at Dickerson Park, 1999]. Thereafter, ongoing maternal care is important, but the calf is not completely dependent upon its mother for nutrients. Sukumar [2003] suggests the maternal contribution is not necessary for survival beyond 2 years, but could be important for maintaining growth rates, body condition, and ultimate reproductive ability of offspring. ACKNOWLEDGMENTS We would like to thank the Dickerson Park Zoo management and entire elephant staff for their cooperation with this study (particularly John Bradford and Stephani Luedde for their assistance). We would also like to thank Michael Jakubasz, the laboratory manager at the Smithsonian National Zoo’s nutrition lab, for all of his assistance in training, troubleshooting and locating, and ordering supplies. We also thank Andrea Drucik, Julia Behler, and Moira McNulty for all of their assistance in helping to process the samples. And last but not least, we acknowledge the cooperation of the well-trained elephants at Dickerson Park Zoo who allowed us to safely collect the milk. All policies and procedures used in this study were approved by the Management of Dickerson Park Zoo and meet the guidelines of AZA Standards for Elephant Management and Care. Zoo Biology 298 Abbondanza et al. REFERENCES Andrews J, Mecklenborg A, Bercovitch FB. 2005. Milk intake and development in a newborn captive African elephant (Loxodonta africana). Zoo Biol 24:275–281. Bernhart, FW 1961. Correlation between growth rate of the suckling of various species and the percentage of total calories from protein in the milk. Nature 191:358–360. Blaxter K. 1989. Energy metabolism in animals and man. Cambridge, UK: Cambridge University Press. Doremus CA. 1881. On the composition of elephant’s milk. J Am Chem Soc 3:55–59. Dubois M, Gilles KA, Hamilton JK, Rebers PA, Smith F. 1956. Colormetric method for determination of sugars and related substances. Anal Chem 28:350–356. Eltringham SK, editor. 1997. The illustrated encyclopedia of elephants. London: Salamander Books. Goto K, Fukuda K, Senda A, Saito T, Kimura K, Glander KE, Hinde K, Dittus W, Milligan LA, Power ML, Oftedal OT, Urashima T. 2010. Chemical characterization of oligosaccharides in the milk of six species of New and Old world monkeys. Glycoconjugate J 27:703–715. Hinde K, Power ML, Oftedal OT. 2009. Rhesus macaque milk: magnitude, sources, and consequences of individual variation over lactation. Am J Phys Anthropol 138:148–157. Jenness R, Sloan RE. 1970. The composition of milks of various species: a review. Dairy Sci Abs 32:599–612. Jensen R, editor. 1995. Handbook of milk composition. New York: Academic Press. Jones DR. 1931. Factors for converting percentages of nitrogen in foods and feeds into percentages of proteins. Circular No. 183. Washington DC: US Department of Agriculture. Kunz C, Rudloff S, Schad W, Brown D. 1999. Lactose-derived oligosaccharides in the milk of elephants: comparison with human milk. Brit J Nutr 82:391–399. Laws RM, Parker ISC. 1968. Recent studies on elephant populations in East Africa. Symp Zool Soc Lond. 21:319–359. Lee PC, Moss CJ. 1986. Early maternal investment in male and female African elephant calves. Behav Ecol Sociobiol 18:353–361. Lemay DG, Lynn DJ, Martin WF, Neville MC, Casey TM, Rincon G, Kriventseva EV, Barris WC, Hinrichs AS, Molenaar AJ, Pollard KS, Maqbool NJ, Singh K, Murney R, Zdobnov EM, Tellam RL, Medrano JF, German JB, Rijnkels M. 2009. The bovine lactation genome: insights into the evolution of mammalian milk. Genome Biol 10:R43. DOI: 10.1186/gb-2009-10-4-r43. Mainka SA, Cooper RM, Black SR, Dierenfeld ES. 1994. Asian elephant (Elephas maximus) milk composition during the first 280 days of lactation. Zoo Biol 13:389–393. Marier JR, Boulet M. 1959. Direct analysis of lactose in milk and serum. J Dairy Sci 42:1390–1391. McCullagh KG, Widdowson EM. 1970. The milk of the African elephant. Brit J Nutr 24:109–117. McKay, GM. 1973. Behavior and ecology of the Asiatic elephant in southeastern Ceylon. Sm C Zoo 125:113. Milligan LA, Gibson SV, Williams LE, Power ML. 2008. The composition of milk from Bolivian squirrel monkeys (Saimiri boliviensis boliviensis). Am J Primatol 70:35–43. Oftedal OT. 1984. Milk composition, milk yield, and energy output as peak lactation: a comparative review. Symp Zool Soc Lond 51:33–85. Oftedal OT. 1986. Growth rate and milk composition: a critical appraisal. In: Filer, LJ, Fomon, SJ, editors. The breast-fed infant: a model for Zoo Biology performance. Report of the 91st Ross Conference on Pediatric research. Columbus, OH: Ross Laboratories. p 50–58. Oftedal OT, Iverson SJ. 1995. Phylogenetic variation in the gross composition of milks. In: Jensen, R, editor. Handbook of milk composition, New York: Academic Press. p 749–789. Oftedal OT, Baer DJ, Allen ME. 1996. The feeding and nutrition of herbivores. In: Kleiman DG, Allen ME, Thomson KV, Lumpkin S, editors. Wild mammals in captivity. Chicago: University of Chicago Press. p 129–138. Oftedal OT. 2012. The evolution of milk secretion and its ancient origins. Animal 6:355–368. Ochs A, Hildebrandt TB, Hentschke J, Lange A. 2001. Birth and hand rearing of an Asian elephant (Elephas maximus) at Berlin Zoo–Veterinary Experiences. Verh Ber Erkrg Zootiere 40:147–156. Osthoff G, De Waal HO, Hugo A, De Wit M, Botes P. 2005. Milk composition of a free ranging African elephant (Loxondonta africana) cow during early lactation. Comp Biochem Phys, Part A 141:223–229. Peters JM, Maier R, Hawthorne BE, Storvick C. 1972. Composition and nutrient content of Asian elephant (Elephas maximus) milk. J Mammal 53:717–724. Poole J. 1997. Elephants. Stillwater, MN: Voyager Press. Power ML, Oftedal OT, Tardif SD. 2002. Does the milk of callitrichid monkeys differ from that of larger anthropoids? Am J Primatol 56:117–127. Power ML, Verona CE, Ruiz-Miranda C, Oftedal OT. 2008. The composition of milk from free-living common marmosets (Callithrix jacchus) in Brazil. Am J Primatol 70:78–83. Powers GF. 1933. The alleged correlation between the rate of growth of the suckling and the composition of the milk of the species. J Pediatr 3:201–216. Ridgway S, Kamolnick T, Reddy M, Curry C. 1995. Orphan-induced lactation in Tursiops and analysis of collected milk. Mar Mammal Sci 11:172–182. Sheldrick D. 1990. Raising baby orphaned elephants: part II. SWARA, 13:23–27. Shennan DB, Peaker M. 2000. Transport of milk constituents by the mammary gland. Physiol Rev 80:925–951. Shoshani J. 1992. Elephants: majestic creatures of the wild. Emmaus, PA: Rodale Press. Simon KJ. 1959. Preliminary studies on composition of milk of Indian elephants. Indian Vet J 36:500–503. Sukumar R. 2003. The living elephants: evolutionary ecology, behavior, and conservation. New York: Oxford Univ. Press. Uemura Y, Asakuma S, Yon L, Saito T, Fukuda K, Arai I, Urashima T. 2006. Structural determination of the oligosaccharides in the milk of an Asian elephant (Elephas maximus). Comp Biochem Physio A145:468–78. Ullrey DE, Crissey SD, Hintz HF. 1997. Elephants: nutrition and husbandry. AAZA nutrition advisory group handbook. Factsheet 004. 20 pp. Available at: http://wildpro.twycross-zoo.org/000ADOBES/Elephants/ D297nut rdietEle_NAG.pdf. Urashima T, Saito T, Nakamura T, Messer M. 2001. Oligosaccharides of milk and colostrum in non-human mammals. Glycoconjugate J 18:357–371. Urashima T, Odaka G, Asakuma S, Uemura Y, Goto K, Senda A, Leo F, Saito T, Fukuda K, Messer M, Oftedal OT. 2009. Chemical characterization of oligosaccharides in chimpanzee, bonobo, gorilla, orangutan, and siamang milk or colostrum. Glycobiology 19:499–508. Welsh U, Fuerhake F, Van Arde R, Buchheim W, Patton S. 1998. Histo- and cytophysiology of the lactating mammary gland of the African elephant (Loxondonta africana). Cell Tiss Res 294:485–501.