Twenty-six species of homeothermic animals were selected as shown in Figure 1. These animals were grouped by habitat into aquatic (n = 8) and terrestrial (n = 18) animals, and by food into carnivorous, herbivorous, and

omnivorous animals. Previously published values were used to compare body temperatures, erythrocyte numbers, and lymphocyte ratios between species [7] - [11] , and mean values were compared between groups. The mean and S.D. were obtained from a numbers of animals (n > 10) except rare species of animals (n = 4, whale and panda) with multiple samples in the range between 19 - 184.

The significance of between-group differences in mean values was assessed by the Mann-Whitney test or one- way analysis of variance followed by post hoc comparisons. The statistical software used was Statcel 2 (OMS Publishing, Saitama, Japan), which is written in Visual Basic for Applications, and IBM SPSS Statistics for Windows version 20.0 (IBM Corp., Armonk, NY). The level of significance was set at p < 0.05 and all data are presented as means ± SD.

The significance of differences between the aquatic and terrestrial species studied was assessed using the Mann- Whitney test (Figure 5). Mean body temperature was significantly lower in the aquatic animals than in the terrestrial animals (36.9˚C ± 0.8˚C vs 38.5˚C ± 1.0˚C; p = 0.002). The number of erythrocytes and the lymphocyte percentages were also lower in the aquatic species than in the terrestrial species (erythrocytes: 4.5 ± 1.3 vs 7.8 ± 3.2 (×10⁶/μL), respectively; p = 0.004; lymphocytes: 21.7% ± 6.4% vs 52.6% ± 16.9%, respectively; p < 0.001) (Figure 6).

The significance of differences between carnivorous, herbivorous, and omnivorous animals was assessed using one-way analysis of variance followed by post hoc comparisons. The carnivorous group had a significantly lowermean body temperature than the herbivorous group (carnivore, herbivore, and omnivore: 37.1˚C ± 0.9˚C, 38.9˚C ± 1.2˚C, and 38.0˚C ± 0.7˚C, respectively; p < 0.01). Carnivores also had significantly fewer erythrocytes than herbivores (carnivore, herbivore, and omnivore: 4.7 ± 1.4, 8.7 ± 4.2, and 7.0 ± 1.7 (×10⁶/μL), respectively; p < 0.01), and lower lymphocyte ratios than the animals in either of the other dietary groups (carnivore, herbivore, omnivore: 41.8% ± 16.7%, 50.2% ± 16.4% and, 77.1% ± 6.9%, respectively; p < 0.01).

Carnivores had lower body temperatures and erythrocyte numbers than herbivores, and lower lymphocyte ratios

than both herbivores and omnivores. Carnivorous animals are hunters that eat the meat of prey animals. Meat contains protein and fat, which are absorbed as amino acids and fatty acids and can be used to create energy. However, meat may also contain foreign antigens, such as bacteria, which may be one reason that the carnivores in this study had low lymphocyte ratios. Leukocytes can be broadly classified as either lymphocytes or granulocytes (with very few monocytes). Therefore, when the lymphocyte ratio is low, the granulocyte ratio is high. Granulocytes serve to protect against bacterial infection. Thus, a low lymphocyte/high granulocyte ratio would help carnivores defend against potentially pathogenic bacteria in their food.

To maintain a stable internal body temperature requires energy; however, carnivorous animals cannot always obtain meat. Therefore, they must avoid unnecessary energy expenditure. Their low body temperatures may be an energy conservation strategy because hypothermia decreases metabolic rate as well as their activity. Other energy-saving strategies include hibernation in winter and heterothermy, such as is utilized by bats. In many bat species, body temperature and metabolic rate are elevated only during activity. When at rest, these animals drastically reduce their metabolic rate and body temperature. They can switch their strategies between hyperthermia (fly) and hypothermia (rest).

In contrast, herbivorous animals have higher body temperatures than carnivores. Herbivores obtain energy from plant materials. However, their digestive organs, especially those of ruminants (such as cattle, goats, and sheep) are unable to produce the enzymes required to break down the plant cell walls (cellulose). Therefore, herbivores have developed a symbiotic relationship with microbes that digest the plant materials for them within a suitable temperature range. The heat generated by microbial fermentation within their gastrointestinal tracts might be one reason for the higher body temperatures of herbivorous animals.

Fifty million years ago the ancestors of modern mammals emerged from the water and survived on land through adaptation to their new circumstance. However, 30 million years ago, some cloven-hoofed animals including the ancestors of modern whales, returned to the water to live as aquatic mammals. Thus, aquatic mammals had to adapt twice to a new external environment and can be said to have experienced a second adaptation event [12] .

Our investigation revealed that homeothermic aquatic animals have lower body temperatures and lymphocyte ratios than homeothermic land animals. A recent study revealed that the panda also has a history of survival and a second adaptation event. In ancient days giant pandas were carnivorous; however, they began to eat bamboo, a food for which their alimentary tract is poorly adapted. The panda coped with this by achieving exceptionally low energy expenditure through reducing the sizes of several vital organs and through maintaining a very low physical activity level and body temperature [13] . Thus, animals with low body temperatures may have undergone greater adaptations, or more adaptation events, than those with higher body temperatures. Early human ancestors began bipedal walking 3.66 million years ago [14] , likely because bipedal walking requires 75% less energy than quadrupedal walking [15] . Bipedal walking became the defining feature of the earliest hominids, and undoubtedly represented a drastic lifestyle change through adaptation; therefore, the low body temperature of humans might be due to our adaptation to bipedal locomotion.

Blood must be maintained at a physiologically appropriate viscosity while performing numerous functions. The high number of erythrocytes in the blood of herbivorous land animals allows rapid clot formation. This may be beneficial for terrestrial prey animals when they are attacked by a predator. By minimizing bleeding they maximize their chance of survival. Additionally, capillary blood flow can be regulated to prevent blood loss and to moderate mitochondrial glycolysis pathways for energy production [16] . A herbivore on land could escape from a predator using prompt energy production via glycolysis. However, due to its low body temperature, an aquatic animal would not benefit from a high number of erythrocytes as much as a land animal. At low temperatures blood becomes more viscous and clots more readily [17] ; therefore, aquatic animals must have fewer erythrocytes than land animals to maintain proper circulatory function.

In this study, aquatic animals had lower lymphocyte ratios than terrestrial animals. This can be explained by the fact that all of the aquatic animals in this study were carnivores. Another possible explanation for this result may be found in their evolutionary process. Aquatic animals have a history of double adaptation―coming onto land 50 million years ago and then returning to the water 20 million years later.

Adaptive responses are often induced by stress and we have previously reported an adaptive glycolytic response to adverse internal conditions [5] [6] . In addition, we have shown that both hypothermia and lymphocytopenia are induced by stress [3] [4] . These findings suggest that carnivorous aquatic animals might have undergone huge stress during the evolutionary process.

This hypothesis may provide insights into our own health. Hypothermia and lymphocytopenia, which are features of aquatic mammals with stressful evolutionary histories, are also characteristics of the internal environment of cancer patients [5] [6] . In Japanese people, the average axillary body temperature had decreased from 36.8˚C to 36.2˚C - 36.3˚C during the last five decades in conjunction with an enormous change in lifestyle. At the same time, the number of cancer patients has significantly increased from 87,895 in 1958 to 368,103 in 2014 [18] .

The results of our investigation suggest that a drastic change in lifestyle that induces excessive adaptation might be a cause of disease. However, our study had a major limitation in that we mainly analyzed previously reported observations; therefore, further research (e.g., larger-scale studies) is needed to support our hypothesis.