蟻蛉幼蟲的覓食行為

Abstract

蟻蛉幼蟲（即蟻獅）是會構築陷阱來捕捉獵物的昆蟲。至今與蟻獅構築沙阱行為相關之研究包羅萬象，但隱含於其行為背後的決策機制仍鮮少被探究。本篇論文分為三個章節，第一章包含蟻獅的生活史簡述與研究對象臀腋蟻蛉（Myrmeleon persimilis）的基本介紹。第二及第三章分別由兩個層面探討蟻獅構築沙阱之行為機制。 第一個探討層面即蟻獅覓食策略的轉換機制。觀察發現，蟻獅採取兩種不同的覓食策略，即構築沙阱覓食（pit-trapping strategy）與伏擊獵物（ambush strategy）。然而兩策略間轉換的機制目前尚未被暸解透徹。在第二章中我運用動態最適化模型（dynamic optimization model），在模擬的各種生態環境（ecological conditions）變化下發現：蟻獅於兩種覓食方式間轉換之行為是最適覓食策略。該模型預測蟻獅在自身能量較高時應採取伏擊策略、能量較低時應構築沙阱來覓食。以上預測與一重要假設相關，即蟻獅在決定覓食策略時面臨生態上的取捨（ecological tradeoff）：構築沙阱一方面能增加成功捕捉獵物的機率，但另一方面（因沙阱所提供的線索）也增加被天敵攻擊的機會。透過實驗，該模型的預測和重要假設均被驗證。綜合各項結果：蟻獅取捨於構築沙阱及伏擊的利與弊，動態地選擇將其適存值（fitness）最大化的策略。 第二個探討層面為蟻獅沙阱的空間分布。野外觀察發現，蟻獅的沙阱常聚集在一起，然而該現象的缺點顯而易見（如易被天敵發現及種內競爭），因此蟻獅的聚集現象仍為生態學上一未解的難題。第三章中，我以賽局理論（game theory）的觀點，提供了對於蟻獅聚集現象的可能解釋。運用空間直觀性個體基準模型（spatially explicit individual based model），我建構出一模擬的空間，該空間由兩隻蟻獅個體以及提供牠們覓食的生態環境所組成。結果指出：當族群中的蟻獅個體採取一演化穩定策略（evolutionarily stable strategy）—和其他個體相鄰時降低換位（relocation）機率，將促使聚集現象形成。為測試由模型得到的結果，我以擲沙粒行為模擬鄰近個體的存在，並設計實驗來檢測蟻獅是否會表現出上述之演化穩定策略。實驗發現，接收到擲沙粒訊息的個體降低了牠們換位的機率。綜合各項結果：因鄰近個體能提供環境中可利用資源（如獵物）的重要訊息，在演化作用下，當與其他個體相鄰時減少換位為一穩定策略，蟻獅個體因而聚集在一起。

Antlions are trap-building insects, but their foraging behavior does not simply consist of the pit construction. To date, a number of aspects of the pit-trapping behavior in antlions have been examined; however, the decision-making mechanisms underlying the observed behavioral responses remain largely unknown. In this thesis, I examined two aspects of the pit-foraging behavior of antlions. A common antlion species in Taiwan, Myrmeleon persimilis, was used as the model organism. In Chapter 1, brief descriptions of the natural history of antlions and the focal species are presented. The first aspect of the pit-foraging behavior examined in this thesis is the act of pit foraging itself. That is, although antlions are typically considered to forge with pits, they can also ambush prey without using pits. Why antlions switch between the two strategies is not fully understood. In Chapter 2, using a dynamic optimization model, I show that the strategy-switching behavior is the optimal foraging strategy under a variety of ecological conditions. In particular, the model predicts that antlions should exhibit the pit-trapping strategy when their energy status is low and should use the ambush strategy when their energy status is high. One of the key assumptions leading to this result is an ecological tradeoff associated with the pit-foraging strategy where pit-foraging increases prey capture success but also increases predation risk. The prediction and the assumption of the model were empirically verified. These results suggest that antlions dynamically choose their strategies to maximize fitness by balancing the cost and benefit of the pit-trapping vs. ambush strategies. The second aspect concerns a spatial aspect of pit-foraging. Antlion pits are commonly found in aggregations in the field, but finding disadvantages of the aggregation (e.g., predation risk and competition) is easier than finding an advantage of it. Thus the pit aggregation is an ecological conundrum. In Chapter 3, I offer an explanation for the aggregation behavior from a game theoretical point of view. By using a spatially explicit individual based model, an ecological scenario where two antlions forage in the common environment was simulated, and how different relocation strategies affect their fitness was examined. The results confirmed that a strategy leading to aggregations (i.e., relocate less when other individuals are nearby) can be the unique evolutionarily stable strategy. To validate the theoretical result, whether antlions follow the evolutionarily stable strategy was empirically examined. The presence of neighbors was simulated by sand tossing, and antlions that received the simulated sand tossing decreased their tendency to relocate. These results suggest that antlions aggregate because the presence of neighbors contains important information about the site quality, and the behavior has been selected through the evolutionary game.