Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2026 Mar 30;45(4):382–390. doi: 10.1002/zoo.70054

A Comparison of Milk Macronutrient Composition From Domestic and Exotic Species Classified in the Order Perissodactyla

Jenna C P Wagner 1,✉, Benjamin Po 2, Mark S Edwards 3, Michael L Power 4, Shweta Trivedi 1, Larry J Minter 5, Kimberly Ange‐van Heugten 1
PMCID: PMC13446449  PMID: 41910413

ABSTRACT

Archived and donated perissodactyla milk samples (n = 209) from 8 domestic horses (Equus caballus), 13 rhinoceroses (representing three species: Diceros bicornis, Ceratotherium simum, Rhinoceros unicornis), 5 tapirs (representing two species: Tapirus bairdii, Tapirus indicus), and 16 wild equids (representing 6 species: Equus ferus przewalskii, Equus grevyi, Equus quagga, Equus zebra, Equus africanus somaliensis, Equus asinus) were collected from different institutions (n = 16). These milk samples ranged from 3 to 200 days post parturition. Milk samples were analyzed at the Smithsonian's National Zoo and Conservation Biology Institute Nutrition Laboratory for major macronutrients: ash, crude protein (CP), dry matter (DM), crude fat, sugar, and gross energy (GE). Analysis of variance (ANOVA) and Tukey post hoc tests were used to compare milk values among both Perissodactyla species and families (Equidae, Rhinocerotidae, and Tapiridae). Overall, there was little variation of milk macronutrient compositions within perissodactyla families besides the differences between black and white rhinos observed (black rhinoceros: CP = 2.81%, fat = 2.10%, sugar = 6.06%, ash = 0.36%, DM = 11.73%, GE = 0.59 kcal/g, white rhinoceros: CP = 1.16%, fat = 0.42%, sugar = 6.47%, ash = 0.23%, DM = 8.96%, GE = 0.36 kcal/g); however, there were significant differences when comparing among families. Large differences (p < 0.05) in the CP content between Tapiridae and Equidae milk were observed (tapir: CP = 5.93%, fat = 2.57%, sugar = 4.62%, ash = 0.68%, DM = 15.09%, GE = 0.75 kcal/g, domestic horse: CP = 2.25%, fat = 1.22%, sugar = 6.35%, ash = 0.38%, DM = 11.04%, GE = 0.50 kcal/g). These findings indicate that domestic equine milk is an insufficient substitute for tapirs, but could be used for rhinoceroses. This result is likely to have a high impact on emerging Perissodactyla breeding and conservation programs with a strong need for milk formulas and replacements.

Keywords: equids, milk, phylogeny, rhinoceros, tapir

Milk samples (n = 209) from 12 different species representing 3 families of Perissodactyla were collected between 3 and 200 days post parturition. Samples were analyzed for (% dry matter) protein, sugar, fat, and ash. Statistical analysis comparing Perissodactyla families indicated that equids and rhinos had similar high‐sugar milks, whereas tapirs had significantly lower sugar, and higher protein and fat.

graphic file with name ZOO-45-382-g002.jpg

1. Introduction

Milk is essential for the survival of mammalian neonates, serving as an evolutionary adaptation that supplies the proteins, sugars, fats, vitamins, and minerals required for growth and development. Most institutions that manage breeding mammals have established milk replacement protocols to support neonates who require supplemental nutrition (Kuo et al. 2011). Scenarios that require additional milk replacer include orphaned or rejected neonates, those that are sick, or instances where the dam is unable to produce an adequate volume of milk or is sick herself (Spaulding 2011). In exotic animal husbandry, care handbooks for various mammal species recommend at least one milk replacer option. These formulas are especially vital in conservation and breeding programs, where they play a key role in supporting the health and population of neonates for endangered and at‐risk species (Association of Zoos & Aquariums Animal Care Manuals n.d.). However, significant differences exist in nutrient composition of milk across mammalian species, and exotic mammal formulas are often dependent on a milk replacer from a small subset of domesticated species whose macronutrient composition may not be suitable (Skibiel et al. 2013).

The order Perissodactyla has 13 of its 18 species listed as vulnerable, endangered, or critically endangered by the International Union for Conservation of Nature (IUCN) Red List (The IUCN Red List of Threatened Species n.d.). Perissodactyla comprise the families Equidae, Tapiridae, and Rhinocerotidae, which are all odd toed ungulates and hindgut fermenters (Price and Bininda‐Emonds 2009). The domestic horse, a member of the Equidae family, is the most well‐known Perissodactyla, commonly kept as a companion animal (Crossman and Walsh 2011). While the milk composition of domestic horses has been analyzed (Musaev et al. 2021), only one study compared the milk macronutrient composition on other Perissodactyla species (Osthoff et al. 2021). Milk supplemental formulation recommendations intended for neonates of Perissodactyla species such as rhinos and tapirs vary widely, with formulas typically falling between the nutrient composition of horse and goat milk (AZA Ungulates n.d.; Versteege 2018; Hall 2008). Without a comprehensive comparison of the milk macronutrient composition across Perissodactyla species, it is challenging to determine the most appropriate milk replacer match for any Perissodactyla species. It is important to note that feeding incorrect milk formulas can result in malnutrition, inappropriate growth, intestinal upset, and even death for neonates (Altazan et al. 2019). For conservation and breeding efforts to succeed, it is crucial to develop effective, species‐specific milk formulas when needed.

This study analyzed the milk macronutrient composition of 12 of the 18 Perissodactyla species and compared the composition within species and across the 3 Perissodactyla taxonomic families. The goal was to identify potential differences in milk macronutrient composition and determine which of these should be further explored to improve the development of Perissodactyla neonate milk replacers for successful conservation and breeding programs.

2. Materials and Methods

2.1. Animals

Milk samples from Perissodactyla species were obtained from the Smithsonian Milk Repository archive at the Smithsonian's National Zoo and Conservation Biology Institute (NZCBI, Washington, DC, USA). Additionally, milk samples were recruited from managed facilities across the United States reported to have archived Perissodactyla milk samples and those with potential Perissodactyla breeding programs (Table 1). The study protocol was approved by North Carolina State University Animal Care and Use Committee (Approval Number, 22‐133; Approval Date, Febuary 4, 2022) and the California Polytechnic State University Animal Care and Use Committee (Approval Number, 2202; Approval Date, April 6, 2022) for studies involving new equine samples. All participating institutions were provided milking collection protocols and acquisition forms to track neonate date of birth, sample date, location, and any other notes on the female or calf health. Samples were included in this research project if the milk was collected between 3 and 200 days post parturition (DPP).

Table 1.

Perissodactyla milk sample metadata. a, , b

Females Institutions Years DPP N
Equidae 24 8 1995–2022 4–176 113
Miniature ass (Equus asinus) 2 1 2003 18–56 2
Somali wild ass (Equus africanus) 5 1 2022 129–176 5
Domestic horse (Equus caballus) 8 2 2022 4–154 90
Przewalski horse (Equus ferus przewalskii) 1 1 2018 4–87 8
Grevy's zebra (Equus grevyi) 6 2 2022 54–91 6
Mountain zebra (Equus zebra) 1 1 1996 16 1
Plains zebra (Equus quagga) 1 1 1995 31 1
Rhinocerotidae 14 11 1985–2022 3–199 73
Black rhinoceros (Diceros bicornis) 6 6 1985–2020 3–163 20
Greater one‐horned rhinoceros (Rhinoceros unicornis) 3 3 2012–2022 2–199 20
White rhinoceros (Ceratotherium simum) 5 4 2008–2022 35–195 33
Tapiridae 5 5 1981–2020 3–103 23
Baird's tapir (Tapirus bairdii) 3 3 1992–2020 3–103 18
Malayan tapir (Tapirus indicus) 2 2 1981–1999 6–101 5
Open in a new tab
a

Headings can be interpreted as the following: number of females (Females), number of institutions housing the animals (Institutions), and number of total milk samples (N) analyzed. The collection date ranges (Years) and ranges of total days post parturition (DPP).

b

Samples were donated from 16 total institutions: Busch Gardens (Tampa, FL, USA), California Polytechnic State University (San Luis Obispo, CA, USA), Denver Zoo (Denver, CO, USA), Detroit Zoo (Royal Oak, MI, USA), Disney's Animal Kingdom (Lake Buena Vista, FL, USA), Fort Worth Zoo (Fort Worth, TX, USA), Franklin Park Zoo (Boston, MA, USA), Hogle Zoo (Salt Lake City, UT, USA), Miami Metro Zoo (Miami, FL, USA), Milwaukee Zoo (Milwaukee, WI, USA), North Carolina State University Equine Unit (Raleigh, NC, USA), Potter Park Zoo (Lansing, MI, USA), Sedgwick County Zoo (Wichita, KS, USA), Smithsonian National Zoo and Conservation Biology Institute (Washington DC, USA), St. Louis Zoo (St. Louis, MO, USA), and White Oak Conservation (Yulee, FL, USA).

The oldest Perissodactyla milk in this study was collected in 1981. The median collection year of samples used (based on female parturition dates) was 2021, with an average of 2011. Of the total samples, 113 samples were from the family Equidae, 73 were from the family Rhinocerotidae, and 23 were from the family Tapiridae. In total, 42 females were sampled across 16 unique institutions. Equids had the most individual females (n = 24), while the rhinos were represented by the highest number of institutions (n = 11). The family Tapiridae was represented with the lowest diversity with 5 females, 5 institutions, and 23 samples (Table 1). The species represented in the analysis included: domestic horses (Equus caballus), black rhinoceros (Diceros bicornis), white rhinoceros (Ceratotherium simum), greater one‐horned rhinoceros (Rhinoceros unicornis), Baird's tapir (Tapirus bairdii), Malayan tapir (Tapirus indicus), Przewalski horse (Equus ferus przewalskii), Grevy zebra (Equus grevyi), plains zebra (Equus quagga), mountain zebra (Equus zebra), Somali wild ass (Equus africanus somaliensis), and miniature donkey (Equus asinus). There was a single milk sample from a mountain zebra (DPP = 16) and a single sample from a plains zebra (DPP = 31). For statistical analyses, we grouped all zebra species together (n = 8, DPP = 16–91).

2.2. Milk Collection

Facilities were provided with standardized milk collection protocols and acquisition forms developed by the Smithsonian Nutrition Laboratory, which were subject to revisions over the years (Peterson et al. 2004). The animals were trained by their caretakers for this procedure for correlating health checks and exams. No chemical stimulations were used for the purposes of this study. Milk collection procedures involved cleaning the nipple area with water and then drying it prior to collection. Manual expression was performed by applying pressure to the area surrounding the nipple and squeezing firmly toward the nipple. Oxytocin was used for some samples. While it was encouraged to extract milk from a single gland, many samples were pooled by collecting in the same container. The milk was collected into clean, tight, sealed sample tubes or bottles and frozen immediately at a minimum of −5°C.

The milk was archived and frozen at −28.8°C at the NZCBI Milk Repository. The number of times each historically archived sample was thawed to room temperature is unknown, but they were thawed to room temperature a minimum of three times. Upon sample assays, milks were homogenized after thawing by vortexing. None of the samples showed signs of deterioration (lumpiness nor inconsistent coloring). When transported between the collection location and NZCBI, the milk was kept on dry ice or ice packs in insulated packaging to keep milk frozen. A total of 209 milk samples were analyzed and fit the criteria for being taken within 3–200 DPP to avoid bias from colostrum which is typically present within the first 3 DPP.

2.3. Nutrient Analysis

Milk dry matter (DM), fat, crude protein (CP), total sugar (sugar), gross energy (GE), and ash were analyzed at the Smithsonian National Zoo and Conservation Biology Institute Nutrition Laboratory following the lab's standard protocols (Hood et al. 2009). CP was calculated by drying 20 µL milk in a forced‐air drying oven at 100°C and running samples through an elemental analyzer (Model 2400; Perkin Elmer, Waltham, MA, USA) to measure nitrogen (Dumas method) (Hood et al. 2009). CP was determined by multiplying total nitrogen by the conversion factor 6.38 (Mariotti et al. 2008). Fat content was measured with a microfat Rose‐Gottlieb procedure that was modified at the laboratory and has been validated (Hood et al. 2009). Total sugar was measured through a phenol‐sulfuric acid colorimetric procedure (DuBois et al. 1956; Marier and Boulet 1959) using lactose monohydrate standards and read at 490 nm on a microplate reader (Model ELX808; BioTek, Winooski, VT, USA). Ash was determined by placing dried milk in a muffle furnace at 550°C for 8 h. DM was calculated throughout sample analysis when drying the milks while subsampling for CP analysis and ash. The GE (kcal/g milk) was calculated using the formula: GE = (9.11 kcal/g * %fat + 5.86 kcal/g * %CP + 3.95 kcal/g * %sugar)/100. To convert to kJ/g, multiply 4.18 kJ/kcal (Hoffman et al. 1998).

2.4. Statistical Analysis

Using the statistical software R version 4.5 (The R Foundation, RStudio, Boston, MA, USA) (and packages dplyr, ggplot, and multcomp), species and families had their mean, median, and range of macronutrient concentrations assessed for GE, DM, CP, sugar, fat, and ash. The general concentrations of macronutrients were assessed on a fresh milk basis (FMB), dry matter basis (DMB), and gross energy basis (GEB).

Multilinear regression models were created to analyze the change of macronutrient composition over parturition time (DPP) within a family. Pearson's correlations were calculated to assess the relationship between macronutrient composition trends over time (DPP) and how the macronutrients values trends correlate over time.

An analysis of variance (ANOVA) was conducted to compare among species and among families for each macronutrient. Subsequently, a post hoc Tukey's Honest Significant Difference test was conducted based on the proceeding ANOVA. Significant values were considered if the p < 0.05.

3. Results

The following significant differences among species analyzed total means between 3 and 200 DPP for DM, sugar, fat, CP, ash, and GE (ANOVA, p < 0.001) were noted.

For calculated milk GE concentrations (Table 2), the Baird's tapir had significantly higher GE values compared to the Somali wild ass (p = 0.001), domestic horse (p < 0.001), greater one‐horned rhinoceros (p = 0.001), white rhinoceros (p = 0.001), and zebra (p = 0.001). Additionally, the white rhinoceros had a significantly lower GE than the domestic horse (p = 0.01), Przewalski horse (p = 0.04), black rhinoceros (p = 0.001), and Malayan tapir (p = 0.01).

Table 2.

The total averages of milk macronutrients (dry matter (DM), sugar, crude protein (CP), fat, ash, calculated a gross energy (GE)) on a fresh milk (FM), dry matter, and gross energy basis for Perissodactyla species over 200 days post parturition.

Species DM GE Total sugar (%) CP (%) Fat (%) Ash (%)
% kcal/g FM DM GE FM DM GE FM DM GE AF DM
Equidae 11.1 0.5 6.4 57.6 53.1 2.1 18.9 24.4 1.2 10.8 22.5 0.4 2.9
Miniature ass 10.1 0.4A,B,C,D 5.5 55.1A,B 56.2 2.4 23.4B,D 35.6 0.4 3.5A,B,D 8.1 0.1 0.7
Somali wild ass 11.5 0.4A,C,D 5.4 47.0A,B 60.8 1.0 8.8A,D 16.0 0.9 7.6A,B,D 23.2 0.2 1.2
Domestic horse 11.0 0.5D 6.6 59.7B 53.3 2.3 19.9A,B,D 25.1 1.2 10.6A,B,D 21.7 0.4 3.1
Przewalski horse 12.1 0.6A,B,D 5.9 49.6A,B 40.2 2.4 19.9A,B,D 23.9 2.2 19.4B,C 35.8 0.4 2.8
Zebrab 11.1 0.3A,C,D 5.0 45.6A,B 62.3 0.9 8.6A,D 17.5 0.7 6.4A,B,D 20.1 0.2 1.6
Rhinocerotidae 9.9 0.4 6.4 67.8 64 1.8 17.1 22.7 0.9 7.0 13.4 0.3 2.5
Black rhinoceros 11.7 0.6A,B,D 6.1 57.6B 50.7 2.8 24.0A 28.7 2.1 12.1A,C 20.6 0.4 3.0
Greater one‐horned rhinoceros 9.2 0.4A,C,D 7.1 76.9B 71.3 1.5 16.0A,D 22.0 0.5 3.1A,D 6.7 0.2 2.6
White rhinoceros 9.0 0.4C 6.5 72.4B 70.5 1.2 12.9D 19.1 0.4 4.6D 10.4 0.2 2.1
Tapiridae 15.1 0.8 4.6 32.8 23.8 5.9 38.3 45.3 2.6 16.0 31.0 0.7 4.0
Baird's tapir 14.6 0.8B 4.4 33.2A 22.6 6.3 41.1C,E 47.9 2.5 16.5B,C 29.6 0.7 4.5
Malayan tapir 16.7 0.7A,B,D 5.2 31.6A,B 26.7 4.6 28.4B,E 34.9 2.9 14.3A,B,D 34.3 0.7 2.4
Open in a new tab

Note: All were analyzed, differing superscripts (A–E) within each column indicate significance differences amongst species (Tukey's post hoc; p < 0.05).

a

Calculated GE = (9.11 kcal/g * %fat + 5.86 kcal/g * %CP + 3.95 kcal/g * %sugar)/100.

b

Zebra = Grevy's, plains, and mountain zebras combined.

For DM, there was a significant difference across the three families (ANOVA, p < 0.01). Tapirs had the highest DM on average (13.77%), while rhinoceros had the lowest (9.31%) (Table 2). Within the rhinoceros, a significant difference in DM was observed between the white rhinoceros and black rhinoceros (ANOVA, p = 0.005), with black rhinoceros having higher DM content (white rhinoceros = 8.96%, black rhinoceros = 11.73%). There was one difference among equid species in DM, with the Przewalski horse (p = 0.03) having higher DM than the domestic horse. There were three distinct outliers exceeding 20% DM, two of which were assumed to be transitional milk from colostrum due to the early time period.

For milk sugar concentrations (sugar on a DMB) (Table 2), there was a significant difference between Baird's tapir and the following species: domestic horse (p < 0.001), black rhinoceros (p < 0.001), greater one‐horned rhinoceros (p < 0.001), and white rhinoceros (p < 0.001). The sugar concentrations ranged from 5.7% to 91.2% of milk DM with the white rhinoceros having the highest mean proportion of DM accounted for by sugar (74.4%), and Baird's tapir had the lowest sugar mean (33.2%). The Tapiridae had significantly lower sugar concentrations than the Equidae and Rhinocerotidae families with no difference in sugar concentrations between Equidae and Rhinocerotidae families.

For CP milk concentrations (CP on a DMB) (Table 2), there was a significant difference between Baird's tapir and the following species: domestic horse (p = 0.001), black rhinoceros (p = 0.001), greater one‐horned rhinoceros (p < 0.001), white rhinoceros (p = 0.001), Przewalski horse (p = 0.001), miniature ass (p = 0.04), zebra (p = 0.001), and Somali wild ass (p = 0.001). Malayan tapir also exhibited a significant difference from zebra (p = 0.02), greater one‐horned rhinoceros (p = 0.004), white rhinoceros (p = 0.001), and Somali wild ass (p = 0.02). The black rhinoceros had higher CP than the white rhinoceros (p = 0.02), and the domestic horse had more CP than the white rhinoceros (p = 0.004). There was a significant difference among all the families, with Equidae having the lowest CP, followed by Rhinocerotidae, and Tapiridae having the highest.

For milk fat concentrations (fat on a DMB) (Table 2), Baird's tapir had significantly higher concentrations than white rhinoceros and greater one‐horned rhinoceros (p = 0.001). The Przewalski horse milk had significantly more fat than the white rhinoceros (p = 0.01), and the greater one‐horned rhinoceros (p = 0.03). Additionally, white rhinoceros milk contained less fat than the black rhinoceros (p = 0.01).

For milk ash concentrations (ash on a DMB) (Table 2), the milk of the Baird's tapir was significantly greater in ash content than the following species: miniature ass (p = 0.006), Somali wild ass (p = 0.001), domestic horse (p = 0.001), black rhinoceros (p = 0.001), white rhinoceros (p = 0.001), greater one‐horned rhinoceros (p = 0.001), zebra (p = 0.001), and Przewalski horse (p = 0.001). There was also a greater milk ash concentration in Malayan tapir compared to Somali wild ass (p = 0.001), black rhinoceros (p = 0.03), white rhinoceros (p = 0.001), and greater one‐horned rhinoceros (p = 0.001). Domestic horses' milk had higher ash content than white rhinoceroses' milk (p = 0.001).

There was a negative correlation observed between DM and DPP in the Tapiridae and Rhinocerotidae families (Figure 1). This was consistent with fat and CP decreasing as sugar increased for both the Tapiridae family (r = −0.65) and Rhinocerotidae family (r = −0.76) over the duration of lactation on a DMB (Figures 2 and 3). Rhinoceros sugar concentrations increased steadily over time on a DMB (Figure 2). For tapirs, the decreasing trend in CP was accompanied by an increase in sugar concentrations on a DMB, with the trendlines intersecting around 65 DPP, at which point tapir milk transitioned from being high in protein to high in sugar (Figure 3). In contrast, equid milk concentrations on a DMB remained relatively constant over time, showing little change in the proportions of the milk's concentration (Figure 4).

Figure 1.

Figure 1

Open in a new tab

The percent dry matter content over time (days of parturition (DPP)) among Equidae, Rhinocerotidae, and Tapiridae with linear trendlines (ANOVA, p < 0.05).

Figure 2.

Figure 2

Open in a new tab

Family Rhinocerotidae macronutrient (ash, crude protein, fat, and sugar as percentage dry matter basis [% DMB] composition over days post parturition [DPP] [range = 3–200]).

Figure 3.

Figure 3

Open in a new tab

Family Tapiridae macronutrient (ash, crude protein, fat, and sugar as percentage dry matter basis [% DMB] composition over days post parturition [DPP] [range = 3–100]).

Figure 4.

Figure 4

Open in a new tab

Family Equidae macronutrient (ash, crude protein, fat, and sugar as percentage dry matter basis [% DMB] composition over days post parturition [DPP] [range = 3–176]).

Between the families Equidae and Rhinocerotidae, significant differences were observed in CP, ash, and GE, however, there was not a difference in sugar or fat concentrations on a DMB (Table 2). Rhinocerotidae had lower CP compositions (% DMB mean: 17.1 < 18.9), lower GE (kcal/g mean: 0.44 < 0.49), and ash concentrations (2.5% DMB < 2.9% DMB) compared to Equidae (Table 2). When comparing Tapiridae and Equidae, Tapiridae exhibited higher levels of CP, fat, ash, and GE and lower sugar levels. Additionally, Tapiridae milk had higher CP concentrations and lower sugar content than the other families.

The largest energy content contributor in Equidae and Rhinocerotidae milk is sugar, accounting for more than 50% (53.14% for Equidae and 67.84% for Rhinocerotidae). Protein was the second largest contributor to energy content in these taxa, averaging 24.42% for Equidae and 22.66% Rhinocerotidae. Fat on a GE basis is lowest for Equidae and Rhinocerotidae at 22.45% and 13.36%, respectively. Tapiridae milk on a GE basis is more evenly distributed with 45.24% of GE being accounted for by protein, 23.77% by sugar, and 30.99% by fat, with a difference of 21.50% between protein and sugar.

4. Discussion

Significant differences in milk macronutrient compositions were observed among the different families and species within the order Perissodactyla, indicating that a single milk replacer may not be suitable for all species in this group. Notable trends were seen in the changes of milk macronutrients over time. Within the family Equidae, the macronutrient change over time was minimal, the sugar concentration specifically was relatively constant (Figure 4), while CP concentrations showed a slight decline. In contrast, rhinoceros milk demonstrated an increase in sugar concentrations over time with a negative correlation between sugar and CP and fat (Pearson's coefficient = −0.76; Figure 2).

The change of milk composition over time is important to consider when developing a milk replacer. For equids, there is little need for adjustments to milk formula to adapt to the change over time, except for water concentrations, which are often accounted for in milk replacer instructions by adjusting the dilutions of the milk replacer powder (Cymbaluk et al. 1993). However, based on these analyses of rhinoceros milk, there is a clear trend of increasing DMB sugar concentrations, while the CP and fat concentrations decreased. Therefore, when developing a milk replacer for all rhinoceros species, it would be advisable to increase sugar concentrations, while reducing the concentrations of CP and fat in line with the observed trends up to 200 days postpartum. In addition, due to compositional differences between white and black rhinos, species‐specific formulas should be considered.

In tapirs, a more pronounced shift occurs, with CP levels being the highest in the first 50 days postpartum, followed by a rise in sugar concentrations, which become the dominant nutrient by Day 103, while CP and fat decrease. This suggests that a tapir milk replacer would need to shift from being a high CP to a higher sugar and lower fat milk at about 50 DPP. The collection period was not extended for 200 DPP for tapirs, however, when comparing only the first 100 DPP among the families, there was still the same significant differences among the macronutrient composition (p < 0.05). Even with tapir milk developing into a high‐sugar milk over time, it still is a comparatively lower sugar concentrated milk ranging between 20% and 40% sugar on a DMB, whereas both rhinoceros and equid milks range between 50% and 80% sugar.

When evaluating milk composition over time on a GE basis, the proportion of CP in tapir milk remains relatively stable, whereas the proportion of GE from sugar increases to surpass fat GE. The trends in milk macronutrient composition over time in rhinoceros and equid milks, when considered on a DMB, closely mirror those on a GEB. However, an exception was observed in equid milk, where sugar content increases when assessed on a GEB.

At the species level, there were significant differences observed within the family Rhinocerotidae, with black rhinoceros exhibiting higher fat and CP content. They also had the highest GE milk among the rhinoceros species. These nutritional differences may be linked to variations in their diets, with white rhinos and greater one‐horned rhinoceros being fed a grazer diet (Laurie 1982), while black rhinos are fed a browser's diet, which is often higher in energy density (van Lieverloo et al. 2009). Despite these dietary differences, the nutritional diet formulations for all the rhinoceros and Perissodactyla species currently in existence are based on the NRC requirements for horses (AZA Ungulates n.d.). Horse diet ecology is predominantly thought of as grazers, and therefore, the nutritional requirements that we use as standards (including milk nutritional requirements) may not be acceptable for all perissodactyla species.

The tapir milk concentration coincides with previous literature, which report tapir milk as having overall a higher CP, and lower sugar milk compared to the rest of the order Perissodactyla (Van Nieuwenhove et al. 2014). Notably, we also observed a difference in CP levels between family Rhinocerotidae and Equidae, highlighting a key point of interest when comparing milk across Perissodactyla species. The high protein content of tapir milk suggests that tapir calves might grow at a relatively faster rate than rhinoceros or equid calves. When considering milk replacers for tapirs, it may be more suitable to provide a formula that has more CP for at least the early lactation period of 3–50 DPP. Equine milk replacers are not appropriate due to their higher sugar content and lower CP (Cymbaluk et al. 1993). Similarly, goat milk replacer, a common alternative, contains 4.5% fat, 3.6% protein, and 4.3% sugar (Turkmen 2017), is also not a close representation of tapir milk composition. However, domestic sow (Sus scrofa) milk, containing 5.9% protein, 5.4% fat, and 4.6% sugar (Zhang et al. 2018), appears to be the best commonly available domestic milk replacer as it closely resembles tapir milk, particularly during early lactation.

Both pigs and tapirs are categorized together in the American Association of Zoos and Aquariums Taxonomic Advisory Group (TAG), instead of tapirs being with their taxonomic relatives such as rhinoceros or equids (Versteege 2018). The similarity in swine and tapir milk concentrations is not necessarily surprising when considering their overlaps in diet and ecological lifestyles. However, when considering the concentrations of the CP, it is crucial to consider the whey to casein protein ratios in the milks and milk replacers due to the different dietary roles they have for growth and development (Gridneva et al. 2018). While both swine and tapir milks are high in CP, further analysis of the amino acid and protein constituents in tapir is needed for more direct comparison with swine literature. A study indicated that the whey to casein protein ratio in lowland tapirs is 1.6:1, (Van Nieuwenhove et al. 2014) which is very similar to the ratio in domestic horses 1.7:1 (Musaev et al. 2021), whereas sow milk reportedly has a ratio of 0.6:1 (Newport and Henschel 1985). Protein is of particular interest since it is the prominent component of tapir milk during early lactation (3–42 DPP).

Our findings are consistent with some previous studies of Perissodactyla species' milk composition (Osthoff et al. 2021; Van Nieuwenhove et al. 2014; Oftedal and Jenness 1988; Gimmel et al. 2018; Osthoff et al. 2025). However, Van Nieuwenhove et al. (2014) saw similar trends of increasing sugar concentrations in lowland tapir milks, they never observed sugar becoming the primary component in the milk. A recent study by Osthoff et al. (2025) focused on white rhinoceros milk, and not only the macronutritional composition, but also micronutrients and fatty acids. Their study included analysis of nine different minerals. This more focused analysis can greatly aid in developing more accurate milk replacers for white rhinoceros over the period of lactation (Osthoff et al. 2025).

Although this represents the largest study of Perissodactyla milks to date, there is still a need to broaden the sample population to reduce the potential for errors caused by variations in factors such as parity, diet, and environment. Additionally, incorporating more species would provide a more comprehensive assessment of the entire Perissodactyla order. The Perissodactyla species missing from the study (Javan rhinoceros (Rhinoceros sondaicus), Sumatran rhinoceros (Dicerorhinus sumatrensis), mountain tapir (Tapirus pinchaque), lowland tapir (Tapirus terrestris), and Asiatic wild ass (Equus hemionus)) are inherently difficult to get due to their critically endangered statuses and permits required. For example, there are currently 18 Javan rhinos that exist today, and they all reside in a national park in Indonesia (Ellis and Talukdar 2020). However, obtaining samples from rainforest rhinoceros species would be fascinating to compare to tapir species who commonly reside in similar environments.

It should be noted that many of the milks from the Smithsonian repository underwent multiple freezes and refreezes. While these cycles are unlikely to affect macronutritional composition (Yochpaz et al. 2020), it should be considered for future studies if micronutritional composition is conducted since bioactive molecules are susceptible to changes in thaw cycles.

Overall, while there are some similarities in milk composition across families within the order Perissodactyla, significant differences also exist. Tapirs stand out due to their milk having higher protein content, with protein accounting for about 75% more of the milk energy compared to the other families. This suggests that feeding tapir calves equine milk replacers might compromise growth, based on these results. There should be investigations on the differences in GE of the milks to the Perissodactyla species growth rates during nursing periods. However, Equine milk replacers are adequate formulas for wild equids and rhinoceros species given the currently available milk replacer options.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

References

  1. Altazan, A. D. , Gilmore L. A., Guo J., et al. 2019. “Unintentional Error in Formula Preparation and Its Simulated Impact on Infant Weight and Adiposity.” Pediatric Obesity 14, no. 12: e12564. 10.1111/ijpo.12564. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Association of Zoos & Aquariums . n.d. “Animal Care Manuals.” AZA.org. Accessed April 5, 2023. https://www.aza.org/animal-care-manuals.
  3. AZA Ungulates . n.d. “Husbandry Manuals.” AZA Ungulates. Accessed April 5, 2023. http://www.azaungulates.org/husbandry-manuals.
  4. Crossman, G. K. , and Walsh R.. 2011. “The Changing Role of the Horse: From Beast of Burden to Partner in Sport and Recreation.” International Journal of Sport and Society 2, no. 2: 95–110. 10.18848/2152-7857/CGP/v02i02/54066. [DOI] [Google Scholar]
  5. Cymbaluk, N. F. , Smart M. E., Bristol F. M., and Pouteaux V. A.. 1993. “Importance of Milk Replacer Intake and Composition in Rearing Orphan Foals.” Canadian Veterinary Journal 34, no. 8: 479–486. [PMC free article] [PubMed] [Google Scholar]
  6. DuBois, M. , Gilles K. A., Hamilton J. K., Rebers P. A., and Smith F.. 1956. “Colorimetric Method for Determination of Sugars and Related Substances.” Analytical Chemistry 28, no. 3: 350–356. 10.1021/ac60111a017. [DOI] [Google Scholar]
  7. Ellis, S. , and Talukdar B. 2020. Rhinoceros sondaicus. The IUCN Red List of Threatened Species: e.T19495A18493900. 10.2305/IUCN.UK.2020-2.RLTS.T19495A18493900.en. [DOI]
  8. Gimmel, A. , Hoby S., Deillon L., et al. 2018. “Milk Composition of Indian Rhinoceros (Rhinoceros unicornis) and Changes over Lactation.” Journal of Zoo and Wildlife Medicine 49, no. 3: 704–714. 10.1638/2017-0011.1. [DOI] [PubMed] [Google Scholar]
  9. Gridneva, Z. , Tie W. J., Rea A., et al. 2018. “Human Milk Casein and Whey Protein and Infant Body Composition Over the First 12 Months of Lactation.” Nutrients 10, no. 9: 1332. 10.3390/nu10091332. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Hall, M. 2008. The Brazilian Tapir. Resources. Australasian Society of Zoo Keeping. https://aszk.org.au/resources/.
  11. Hoffman, R. M. , Kronfeld D. S., Herbein J. H., Swecker W. S., Cooper W. L., and Harris P. A.. 1998. “Dietary Carbohydrates and Fat Influence Milk Composition and Fatty Acid Profile of Mare's Milk.” Journal of Nutrition 128, no. 12: S2708–S2711. 10.1093/jn/128.12.2708S. [DOI] [PubMed] [Google Scholar]
  12. Hood, W. R. , Voltura M. B., and Oftedal O. T.. 2009. Methods of Measuring Milk Composition and Yield in Small Mammals, 529–553. John Hopkins University Press. [Google Scholar]
  13. Kuo, A. A. , Inkelas M., Slusser W. M., Maidenberg M., and Halfon N.. 2011. “Introduction of Solid Food to Young Infants.” Maternal and Child Health Journal 15, no. 8: 1185–1194. 10.1007/s10995-010-0669-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Laurie, A. 1982. “Behavioural Ecology of the Greater One‐Horned Rhinoceros (Rhinoceros unicornis).” Journal of Zoology 196, no. 3: 307–341. 10.1111/j.1469-7998.1982.tb03506.x. [DOI] [Google Scholar]
  15. Marier, J. R. , and Boulet M.. 1959. “Direct Analysis of Lactose in Milk and Serum.” Journal of Dairy Science 42: 1390–1391. [Google Scholar]
  16. Mariotti, F. , Tomé D., and Mirand P. P.. 2008. “Converting Nitrogen Into Protein—Beyond 6.25 and Jones' Factors.” Critical Reviews in Food Science and Nutrition 48, no. 2: 177–184. 10.1080/10408390701279749. [DOI] [PubMed] [Google Scholar]
  17. Musaev, A. , Sadykova S., Anambayeva A., Saizhanova M., Balkanay G., and Kolbaev M.. 2021. “Mare's Milk: Composition, Properties, and Application in Medicine.” Archives of Razi Institute 76, no. 4: 1125–1135. 10.22092/ari.2021.355834.1725. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Newport, M. J. , and Henschel M. J.. 1985. “Growth, Digestion, and Protein Metabolism in Neonatal Pigs Given Diets Containing Whey as the Predominant or Only Source of Milk Protein.” Journal of Pediatric Gastroenterology and Nutrition 4, no. 4: 639–644. 10.1097/00005176-198508000-00025. [DOI] [PubMed] [Google Scholar]
  19. Oftedal, O. T. , and Jenness R.. 1988. “Interspecies Variation in Milk Composition Among Horses, Zebras and Asses (Perissodactyla: Equidae).” Journal of Dairy Research 55, no. 1: 57–66. 10.1017/s0022029900025851. [DOI] [PubMed] [Google Scholar]
  20. Osthoff, G. , Beukes B., Steyn A. C., et al. 2021. “Milk Composition of White Rhinoceros over Lactation and Comparison With Other Perissodactyla.” Zoo Biology 40, no. 5: 417–428. 10.1002/zoo.21618. [DOI] [PubMed] [Google Scholar]
  21. Osthoff, G. , Hugo A., and Nieuwoudt P.. 2025. “Composition of White Rhinoceros Colostrum and Changes During Early Lactation to Mature Milk.” Zoo Biology 44: 432–441. 10.1002/zoo.21907. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Peterson, A. B. , French K. R., Russek‐Cohen E., and Kohn R. A.. 2004. “Comparison of Analytical Methods and the Influence of Milk Components on Milk Urea Nitrogen Recovery.” Journal of Dairy Science 87, no. 6: 1747–1750. 10.3168/jds.S0022-0302(04)73329-9. [DOI] [PubMed] [Google Scholar]
  23. Price, S. A. , and Bininda‐Emonds O. R. P.. 2009. “A Comprehensive Phylogeny of Extant Horses, Rhinos and Tapirs (Perissodactyla) Through Data Combination.” Zoosystematics and Evolution 85, no. 2: 277–292. 10.1002/zoos.200900005. [DOI] [Google Scholar]
  24. Skibiel, A. L. , Downing L. M., Orr T. J., and Hood W. R.. 2013. “The Evolution of the Nutrient Composition of Mammalian Milks.” Journal of Animal Ecology 82, no. 6: 1254–1264. 10.1111/1365-2656.12095. [DOI] [PubMed] [Google Scholar]
  25. Spaulding, C. E. 2011. The Complete Care of Baby Animals. Skyhorse Publishing. [Google Scholar]
  26. The IUCN Red List of Threatened Species . n.d. “Family: Perissodactyla: Status Statistics.” Accessed April 4, 2023. https://www.iucnredlist.org/search/stats?taxonomies=100194&searchType=species.
  27. Turkmen, N. 2017. “Chapter 35 ‐ The Nutritional Value and Health Benefits of Goat Milk Components.” In Nutrients in Dairy and Their Implications on Health and Disease, edited by Watson R. R., Collier R. J., Preedy V. R., et al., 441–449. Academic Press. 10.1016/B978-0-12-809762-5.00035-8. [DOI] [Google Scholar]
  28. van Lieverloo, R. J. , Schuiling B. F., de Boer W. F., et al. 2009. “A Comparison of Faecal Analysis With Backtracking to Determine the Diet Composition and Species Preference of the Black Rhinoceros (Diceros bicornis minor).” European Journal of Wildlife Research 55, no. 5: 505–515. 10.1007/s10344-009-0264-5. [DOI] [Google Scholar]
  29. Van Nieuwenhove, C. , Pérez M. E., Hernández M., and Luna Pizarro P.. 2014. “Characterization of Gross Composition, Energy Value, and Fatty Acid Profile of Milk From Lowland Tapir (Tapirus terrestris) During Different Lactation Periods.” Canadian Journal of Zoology 92, no. 9: 803–809. 10.1139/cjz-2014-0010. [DOI] [Google Scholar]
  30. Versteege, L. 2018. “EAZA Best Practice Guidelines for the White Rhinoceros (Ceratotherium simum).” In EAZA Tag, 1–82. Beekse Bergen. [Google Scholar]
  31. Yochpaz, S. , Mimouni F. B., Mandel D., Lubetzky R., and Marom R.. 2020. “Effect of Freezing and Thawing on Human Milk Macronutrients and Energy Composition: A Systematic Review and Meta‐Analysis.” Breastfeeding Medicine 15, no. 9: 559–562. [DOI] [PubMed] [Google Scholar]
  32. Zhang, S. , Chen F., Zhang Y., et al. 2018. “Recent Progress of Porcine Milk Components and Mammary Gland Function.” Journal of Animal Science and Biotechnology 9: 77. 10.1186/s40104-018-0291-8. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Articles from Zoo Biology are provided here courtesy of Wiley

RESOURCES