STAGES OF ADAPTATION IN INSECTS 193 



Blumenbesuch v. Insekten a. Freilandpfl. d. bot. Gart. zu Berlin,' Jahrb. kgl. bot. 

 Gart. zu Berlin, iii, 1884; iv, 1886: 'Beitrage zur blutenbiol. Statistik,' Verh. bot. 

 Ver., Berlin, xxxi, 1890: 'Blutenbiol. Floristik,' pp. 386-8, Stuttgart, 1894). Loew's 

 scheme has also been used by MacLeod and by Heinsius for statistical observations, 

 and I myself have employed it in several investigations relating to the statistics of 

 flower pollination, during which the conclusion has more and more forced itself 

 on me that Loew has adopted the right method, though some change may be 

 necessary in certain details. 



Loew's first group — Allotropous Insects — includes Stage I and part of Stage II 

 of VerhoefTs classification. To the second group — Hemitropous Insects — belong 

 Verhoeffs Stages IV, III, and VI. The third group — the Eutropous Insects — 

 corresponds to Verhoeffs Stage V, with a small part of Stage VI. The last of the 

 groups established by Loew — Dystropous Insects — includes species which were not 

 taken into consideration by Verhoeff. 



Loew's stages of adaptation of insects to flower visits are defined as follows : — 



I. Allotropous Insects. Unequally and only slightly adapted flower visitors of 

 little value for pollination. They either lack special structural characters adapting 

 them to flower visits — except such as may be involved in the features common to the 

 group — or if any adaptations of the kind exist they are incipient. In addition to 

 flower-food the insects of this group also nourish themselves on a great variety of 

 other substances (e. g. the social Vespidae, the blood-sucking Empidae, Tabanidae, 

 and others), and some of them are destroyers of flowers (e. g. many beetles, species 

 of Cephus, and others). Their movements while visiting flowers are mostly irregular, 

 only attaining any constancy in the more highly adapted species. Flower forms 

 corresponding to allotropous visitors are very sparingly developed, and by no means 

 to be regarded as adaptations to such insects alone, though these display marked 

 preference for flowers of the kind. 



To this group belong : — 



1. Hymenoptera. Short-tongued digging- wasps (Fossores) and ruby- wasps 

 (Chrysididae), most of the true wasps (Diploptera, e.g. Vespa, Polistes), saw-flies 

 (Tenthredinidae), and ichneumons (Ichneumonidae). 



2. Diptera. Muscidae, Empidae, Tabanidae, Therevidae, Leptidae, Stratio- 

 myidae, Dolichopidae, Bibionidae, and others. 



3. Coleoptera. Dermestidae, Coccinellidae, Nitidulidae, Lamellicornia (in part), 

 Melyridae, Cerambycidae, Lepturinae, Oedemeridae, and others — in fact most 

 beetles, excepting such as are particularly hurtful to flowers. 



4. Neuroptera, Orthoptera, and Hemiptera. Such of these as are occasional 

 flower visitors. 



II. Hemitropous Insects. Partially adapted flower visitors of moderate value 

 for pollination. The adaptations to successful flower visits are always distinctly 

 recognizable, but are much more feebly developed than in the following group. 

 The movements while visiting flowers are skilled, but not so regular and 

 constant as in Group III. Flower forms specially adapted to these insects 

 are rare. 



DAVIS 



