As distinguished from true or natural hosts one must separate what may be termed casual or accidental hosts. All animals which come in contact with one another, or which live in close proximity, may exchange fleas. So even bird-fleas may be collected from mammals and typically mammalian fleas from birds. In this fashion puzzles may arise which tax the ingenuity of the collector to solve. Bird-fleas are sometimes found on bats, and this may be obviously attributed to the bats having inhabited a hole which was tenanted by starlings or an old loft infested with the fleas of pigeons. All beasts of prey are sometimes found to harbour the fleas of animals they have devoured. Rabbits’ fleas are found on wild-cats; hedgehogs’ fleas on foxes; mice fleas on weasels; and fleas characteristic of small birds on stoats. So also in the case of mice, rats and voles with holes and runs in the same hedgerow, the parasites usually peculiar to one are not uncommonly found on the others. It is sometimes difficult to determine the true host of a flea. Much more puzzling to explain are the reasons which confine a flea to a certain host and which cause closely allied hosts to have different fleas. The fleas from the house-martin and the sand-martin are quite different; those from the domestic fowl and the domestic pigeon are distinct species. The causes which have affected the evolution of the various forms of flea are too obscure to enable anyone at the present day to offer any satisfactory explanation. Speaking generally, the fleas found on birds have points in common, and they probably form a natural group to themselves. What may be called true bird-fleas have been collected from almost all European birds. An unwieldy genus (_Ceratophyllus_) comprises many species of different flea. Some species are very abundant and infest the nests of many different birds. Others are extremely rare. One of these rarities (_C. vagabundus_) is found in the nests of puffins and other sea-birds. Another has been collected on antarctic petrels. Penguins have a special genus of flea to themselves. A specimen, unique at one time (_Ceratophyllus borealis_), in Mr N. C. Rothschild’s collection was obtained from the gannet. It has now been found on rock-pipits in the Shetland Islands. Two very rare fleas (_C. farreni_ and C. _rothschildi_) are found in the nests of house-martins; yet the nests of these birds are infested with common species besides. A plague flea (_Xenopsylla_) has been found on an African swift. Forty-six different species of flea have been found in the British Islands, but many of these are extremely scarce.