1. The Small Elephants of Sicily, Only About 1 Meter Tall
The species *Palaeoloxodon falconeri* lived on Sicily with a shoulder height of 0.9 to 1.2 meters (about 3 to 4 feet) and an estimated average weight of 252 kg (about 555 pounds). Its ancestor, *Palaeoloxodon antiquus*, is estimated to have weighed about 11,500 kg (about 25,350 pounds), meaning the small elephant was about 2% of its ancestor’s weight [1]. However, these weights are estimates derived from fossil bones, and they vary between studies. Another skeletal reconstruction suggested that adult males were about 96.5 cm (about 38 inches) tall and weighed about 305 kg (about 672 pounds), while females were about 80 cm (about 31 inches) tall and weighed about 168 kg (about 370 pounds) [2]. You should read these numbers as approximations.
In this article, "insularization" refers specifically to changes in the body size of island animals. The extinction of unique island species is a different topic handled in other notes. Here, we focus only on body size.
2. Why Small Island Elephants Still Grew Slowly
If you think of mice, it seems natural that "small animals grow fast and have short lives." However, a study examining the internal tissue of small elephant bones reported results that contradict this idea. The study estimates that these elephants matured at about 15 years old and lived for at least 68 years [1]. The figure of 68 years is a lower-bound estimate of "at least this long," not the exact lifespan. Even so, it seems that becoming smaller is not necessarily linked to changes in growth rate or lifespan in a fixed way. Does the mechanism for shrinking involve growth speed as well as body size? This question remains open.
3. Foster's 1964 Study of Island Animal Sizes
In 1964, J.B. Foster published a paper comparing 116 island animals to their mainland relatives. He reported that rodents tend to get larger on islands, while rabbits, carnivores, and even-toed ungulates (like deer) tend to get smaller [3]. This tendency later became known as "Foster’s rule" or the "island rule." In 2005, Lomolino defined this as a gradual trend where small species head toward gigantism and large species toward dwarfism, and he examined its relationship with island area and distance from the mainland [4].
Two reasons were suggested: small animals can grow larger if there are few predators, and large animals shrink if food is limited. However, both are hypotheses. The causes have not been proven [3].
4. Yakushima Deer: A Small Island Deer in Japan
A common example in Japan is the Yakushima deer (*Cervus nippon yakushimae*). It is particularly small among Japanese sika deer subspecies, with a body height of about 80 cm (about 31 inches). For mature individuals in western Yakushima, females weigh about 19 to 25 kg (about 42 to 55 pounds), and males weigh about 24 to 37 kg (about 53 to 82 pounds) [5]. Having short limbs relative to their body is also a feature. These figures are for the population in western Yakushima, not the whole island.
Hypotheses for this shrinking include the lack of evidence that medium-to-large carnivores naturally inhabited Yakushima, and competition within the same species for food [6]. These are still at the hypothesis stage. You may also see explanations stating they weigh about half as much as deer on Honshu, but since deer get smaller as you go south on Honshu, it may not be decided by island versus mainland alone. This point has weak sources, so we only mention it as a tendency.
5. Island Animals That Became Larger, Like the Flores Giant Rat
Examples of animals getting larger are also known. On Flores Island in Indonesia, the Flores giant rat still lives today, with a head-and-body length of 41 to 45 cm (about 16 to 18 inches) [7]. This is a much larger body than typical rodents. It is a representative example of Foster’s tendency that "rodents tend to get larger on islands." If small elephants and Yakushima deer are on the "getting smaller" side, this rat is on the "getting larger" side.
Listing examples where body size changes in opposite directions compared to the mainland makes it seem like islands exert some common pressure. However, as the next section shows, researchers do not agree on this.
6. Is Island Body-Size Change a Law or Just a Bias?
In 2021, Benítez-López et al. examined over 2,000 pairs of island and mainland populations, accounting for evolutionary relatedness. They reported that insularization trends are widely seen in mammals, birds, and reptiles, though less clear in amphibians. The strength of the trend varies with climate, island size, and distance from the continent; for mammals and reptiles, it was stronger on smaller, more distant islands [8][10].
In contrast, Meiri et al. concluded in 2008 that large-scale data accounting for evolutionary relatedness showed no evidence of an "island rule." They argued that carnivores, even-toed ungulates, and heteromyid rodents get smaller on islands, while murine rodents get larger, but these are trends specific to each lineage [9]. These two conclusions are divided. You cannot say one is "proven" or the other "disproven" alone. Differences in how data is taken or the criteria for calling something a "law" may influence this.
7. How to Investigate Island Animal Sizes Yourself
What you can do today: Write down the weights of the animals mentioned in this article (small elephant about 252 kg, Yakushima deer about 19 to 37 kg, elephant ancestor about 11,500 kg) on paper and line them up. You will see that even when we say "small," there are big differences between animals.
What you can do when you go out: Look at museum exhibits to see if the body sizes of "ancestors" and "island relatives" are displayed side by side. If the descriptions of fossils or skeletal specimens include a range of estimated weights, you can compare them with the numbers in this article.
When you want to know more: Search for the terms "insularization" and "island rule." Starting with explanations from research institutions or academic societies will help you grasp the overall picture of the debate. These two terms are the entry points to the discussion in this article.