( cxvii ) 



The disti-ibutiou of the Hawk Motlis of North America illustrates a marked 

 fauiiistie division of the Continent into an Atlantic and a Pacific Subregion. 

 The Atlantic Subregion reaches westwards to the Mississijipi plains and includes 

 part of Texas. It contains far more Neotropical iraiuigrauts than the AVestern 

 Subregion, Florida offering special facilities for a northward migration of the 

 Neotropical Sj/liingidae of the "West Indies, none of the Neotropical sj}ecies 

 which occur regularly or (iccasiiiually iu Georgia, Carolina, etc., being absent 

 from <'uba. The genera characteristic for tlie Atlantic Subregion are thirteen: 

 Dolba, Ccratomia, Isogramma, Isoparce, Atreus, Laparit, Ampeloeca, Iktrapsa, 

 Dculamia, Ampliion, Calusi/mholus, and ('rcssoiiia, besides Uphecodina, which 

 occurs, however, also in the Pacific district of the Palaearctic Region. The 

 Atlantic Subregion is, therefore, rich in genera peculiar to itself, while the 

 Pacific side of the (Jontinent, from the Rocky Mountains to the Pacific Ocean, 

 has only three genera restricted to itself (Arcfonotiis, Euproserpinus, and 

 I/icti/osomn). Of these western genera, Arctonotus and Euproserpinug are 

 derivations from Fioserp/nus, and Dictijosoma from tlie Atlantic Ceratomia. 

 The two Pacific species of Proserpinus and the only European one are more 

 specialised than the two s])ecies which belong to the Atlantic Subregion. We 

 have, therefore, the remarkable fact that what is characteristic for the Nearctic 

 Region is essentially Atlantic. The Atlantic Subregion is the birthplace of the 

 Nearctic Sphingid fauna. Tliis is borne out l-.y the two Holarctic genera Sphinx 

 and HaemorrliKc/ia, which have together five species in the Atlantic and only 

 two in the Pacific Subregion of North America. This distribution gives a hint 

 how to explain the glaring contrast existing between temperate and tropical 

 America in the Sesiinae and Clioerocampinae. One would have expected to find 

 that, as is the case in Ac/ierontiinae, the large Neotropical stock of Si'siinae 

 had given rise to a crop of Nearctic specialisations, and that the Ohoerocampine 

 genus Xi/l(ipli<ines, flourishing with fifty species in the Neotro|)ical countries, 

 had some special representatives in temperate North America. Temperature 

 and other atmospheric factors as such cannot be the cause of the deficiency iu 

 the North American fauna, since the Sesiine genus llnemorrhagia is almost 

 exclusively an inhabitant of the northern temperate countries of the Old and 

 New Worlds ; since, secondly, the Palaearctic Region has eight species of Clioero- 

 campinae to itself, besides Perqcsa elpenor, which reaches into India ; since, 

 thirdly, tomi)erate South America has several SesiiiKie and Clioerocampinae which 

 do not go far north into the tro2>ics ; and as, fourthly, the tropical plains as well 

 us the tcmjierate high mountainous districts of South and Central America are 

 inliaiiited l)y Sesiinae and Clioerocampinae. However, if it is hardly possible that 

 conditions of life are at the root of the discrepancy, the paucity of Sesiinae 

 in North America and the absence of any indigenous species of Choeroca npinae 

 from that Region must find an explanation in the geographical history of the 

 Western Hemisphere. The explanation is rendered complicated by the fact 

 that there are no less than seven purely Nearctic genera of Aeherontiinae 

 whicl] are of Neotrojiic^al extraction, and therefore apt to contradict any likely 



