把一条鱼放入一个有着吃不完的食物的鱼缸，它会狼吞虎咽直至体形增大了数倍。这或许是因为鱼有一个比实际需求大得多的消化系统。然而考虑到鱼在野外不可能吃这么多，那么为什么还要花这么大的力气来保持所有这些消化道呢？
根据7月6日《自然》杂志网络版报道，对600种鱼（包括图中的太阳鱼）进行的一项新的分析表明，大的消化道有助于鱼类应对野外的盛宴或饥荒。两到三倍于所需的一个消化系统能够帮助鱼类在发现食物的时候足吃足喝，并储备卡路里以备食物短缺时之需。并且，从长远来看，这使得拖着一肚子内脏变得有了价值。（生物谷Bioon.com）
生物谷推荐原文出处：
Nature doi:10.1038/nature10240
Excess digestive capacity in predators reflects a life of feast and famine
Jonathan B. Armstrong; Daniel E. Schindler
A central challenge for predators is achieving positive energy balance when prey are spatially and temporally heterogeneous. Ecological heterogeneity produces evolutionary trade-offs in the physiological design of predators; this is because the ability to capitalize on pulses of food abundance requires high capacity for food-processing, yet maintaining such capacity imposes energetic costs that are taxing during periods of food scarcity1, 2. Recent advances in physiology show that when variation in foraging opportunities is predictable, animals may adjust energetic trade-offs by rapidly modulating their digestive system to track variation in foraging opportunities1. However, it is increasingly recognized that foraging opportunities for animals are unpredictable3, which should favour animals that maintain a capacity for food-processing that exceeds average levels of consumption (loads)2, 4. Despite this basic principle of quantitative evolutionary design, estimates of digestive load:capacity ratios in wild animals are virtually non-existent1. Here we provide an extensive assessment of load:capacity ratios for the digestive systems of predators in the wild, compiling 639 estimates across 38 species of fish. We found that piscine predators typically maintain the physiological capacity to feed at daily rates 2–3 times higher than what they experience on average. A numerical simulation of the trade-off between food-processing capacity and metabolic cost suggests that the observed level of physiological opportunism is profitable only if predator–prey encounters, and thus predator energy budgets, are far more variable in nature than currently assumed.