TRANSACTIONS OF SECTION D, 733 



majority are more or less widely distributed, and may be placed in at least two 

 great faunistic groups. One group inhabits the lowland waters of a comparatively 

 high temperature, while the other is especially charaoteristic of highland lakes 

 and streams possessing a deeper temperature. Both groups show many interesting- 

 structural modifications in harmony with their natural environment. 



4. On some Arctic and Antarctic Collembola. 

 By Professor George H. Carpenter, B.Sc, M.R.I. A. 



The last ten years have been marked hy great advances in the systematic 

 study of the Collembola or springtails. Collections from many parts of the world 

 have been worked out, but the most striking results have been obtained from the 

 examination of specimens brought from the Arctic and Antarctic regions by 

 various expeditions. The comparative richness of the Collembolan fauna of 

 remote northern and southern lands is remarkable. In the Arctic, Greenland 

 has about 20 species of springtails, Spitzbergen 16, and Franz-Joseph Land 7 ; 

 while in the Antarctic, Kerguelen has 5, Graham Land and the South Shet- 

 lands 4, South Georgia 6, the Falklands 10, and South Victoria Land at least 2. 



■ According to the views of most recent students, the Poduridae and the 

 Isotominae are nearest to the primitive stock of the order, the Entomobryinfe, the 

 Tomocerinae, and the Symphypleona being more highly specialised. It is suggestive 

 to find that in both the Arctic and Antarctic faunas the primitive sections are 

 well repre.'sented, while the specialised genera have but very few species. And in 

 the more remote and insular regions the higher groups seem entirely absent. 



Of much interest is the presence of two Arctic Isotomines in our own islands. 

 These are Agrenia hidenticulata (Tullb.), a species both Arctic and Alpine, dis- 

 covered last year in Irish and North British mountain streams, and Proisotoma 

 Beselsii (Packard), which inhabits the Arctic regions of both the Old and New 

 Worlds and the coast of Scotland. ' Bi-polarity ' in the Collembola is shown by 

 Wahlgren's recent record of this latter species from Terra del Fuego and by the 

 presence of a closely allied form {Proisotoma Brucei, Carp.) on the South Orkneys. 

 Such distribution indicates a high antiquity (probably Mesozoic) for the species. 



A similar conclusion is suggested by a comparison of the distinctively Antarctic 

 springtails. Several genera are apparently confined to the southern regions. 

 Among these Cryptopygus (Willem) is represented by identical or nearly allied 

 species in Terra del Fuego, Graham Land, South Shetland, South Orkneys, and 

 South Georgia. Turning to genera of wider range we find the same Isotoma 

 (7. octo-oculata, Willem) present in Graham Land, South Shetland, South Orkneys, 

 and Kerguelen, while the Isotoma of South Victoria Land (7. klovstadi, Carp.) is 

 closely allied to a Fuegian species. Such distributional facts suggest a con- 

 siderable geological age for the species and a former wide extension of the 

 Antarctic continent. 



The National Antarctic ('Discovery ') Expedition collected from moss at Granite 

 Harbour, South Victoria Land, a remarkable springtail, referable to tbe 



Poduridce, but showing some striking affinities to the Isotomince. This insect 



apparently the most southerly terrestrial animal yet known — will be described 

 .ind figured in the forthcoming part of the Expedition Reports. 



5. On Diaposematism, or the Interchange of Characters between Distaste/til 

 Forms. ByY. A. Dixey, M.A., M.D. 



When in the year 1879 Fritz Muller put forward his theory of common warn- 

 ing colours, or the assimilation of one distasteful form to another for the sake of 

 mutual protection against insectivorous enemies, he recognised the probability, or 

 even certainty, that the approach would not necessarily be one-sided, but mioht 

 be in the strict sense convergent, each form in some respects advancing to m'eet 

 the other. This suggestion, however, so far as F. Miiller was concerned, remained 



