374 BULLETIN 120, UNITED STATES NATIONAL MUSEUM. 



step? Our data seem to give us no solution of this puzzle. The 

 northern Mexican desert has apparently been an important bar to" 

 free passage between the two continents, but Hylidae have readily 

 passed it and toads and frogs have in considerable numbers moved 

 across it to tropical Central America. There have been two regions 

 of hindrance, the Isthmus of Panama and the northern Mexican 

 desert. That the latter should bar the passage of moist-skinned 

 forms like most of the Anura seems but natural, but why has the 

 Isthmus been a bar during the recent times when it has not been 

 interrupted ? 



Temperature conditions can hardly have been the controlling ele- 

 ment. The Pelobatidae are known from tropical Malaysia, but have 

 not entered the American Tropics. This may well be due to the 

 Mexican desert holding them back. The Gastrophrynidae thrive in 

 the Southern Temperate Zone, though they prefer the Tropics. The 

 Mexican desert probably has hindered their migration but has not 

 prevented it altogether, for they are represented in southeastern 

 United States. The Leptodactylidae live in both Tropical and Tem- 

 perate Zones in the South. Their northward movement has prob- 

 ably been prevented only by the Mexican desert. The Dendrohatinae 

 are apparently strongly tropical in their preferences, but they might 

 well have moved up to the northern coast of the Gulf of Mexico and 

 into Florida, except for the desert. But no review of the data, group 

 by group, explains why the Central American forms have not passed 

 south into South America. The axis of the Isthmus is high ground 

 and there is something of a coastal plain most of the way on one or 

 both shores, so that there is altitudinal diversity of climate, and the 

 hindrance can not well be climatic. 



Among the Opalinidae we find no indication that conditions of 

 temperature, except of course extreme cold, have proven a bar to 

 migration. Protoo-palina is cosmopolitan (fig. 239), excepting only 

 New Zealand, Madagascar, and southern Asia, and they may be un- 

 known from southern Asia only because our material from this area 

 has been very scant. ZeZ?meZZa is Antarctic (fig. 248). Its Lepto- 

 dactylid hosts have not entered North America to any extent, and 

 apparently the toads have not passed from tropical America to North 

 America after meeting and adopting Zelleriella in tropical America. 

 Cepedea is cosmopolitan, except for Australasia and New Zealand. 

 (Fig. 249.) Neither temperature or other climatic conditions have 

 hindered its spread. It is doubtful whether Cepedea has reached 

 South America from the north or only from Africa by the northern 

 trans-Atlantic bridge. Its having entered from Africa seems indi- 

 cated by a study of the African and South American species 

 (p. 337 et seq.). The broad OpaZinae, Eastern Hemisphere forms, 





