314 BULLETIN 120, UNITED STATES NATIONAL MUSEUM. 



is barred by the present distribution of the genus Gepedea which is 

 descended from Protoopalina. To this we will refer later (p. 336). 

 Compare the evidence soon to be noted that group 4 of the genus 

 Protoopalina arose in the Jurassic while Australia was connected 

 with Asia-Malaysia but was not connected with Africa and South 

 America (p. 317). The genus Protoopalina^ therefore, seems Equa- 

 torian in origin and early distribution. As this is the most archaic 

 genus of the Opalinidea we have thus an origin of the family in 

 Equatoria, and during the Triassic period or before. 



AccejDting this origin for the genus Protoopalina^ we may say fur- 

 ther that in Jurassic times' (fig. 233) Australia separated from 

 Equatoria and united with Asia-Malaysia which then became popu- 

 lated by immigrant Protoopalinas. 



Group 2 (fig. 240, crosses). 



P. dorsalls (p. 36) . . .in Limnodynastes (Leptodactylidae) Australia. 



P. peronii (p. 37) in Limnodynastes (Leptodactylidae) Australia. 



These two species are considerably flattened, especially in front, 

 and they are lance-shaped, being much less slender posteriorly than 

 the members of the first group of species, having no naked, sharp- 

 pointed, posterior tip. Group 2 seems nearly related to group 1. 

 Its species were probably evolved in Australia, and at a time not 

 indicated. 



Group 3 (fig. 241). 



P. caudata (p. 40) in Discoglossidae Europe. 



P. macrocaudata (p. 49) in Discoglossidae Eastern Asia. 



P. orientalis (p. 48) in Discoglossidae Eastern Asia. 



P. nutti (p. 45) in Rana Tropical Africa. 



P. stevensoni (p. 50) in Bufo Tropical Africa. 



P. australis (p. 68) in Hyla Australia. 



Protoopalina caudata and P. nutti show both slender and stocky 

 forms, the stocky forms having usually an abrupt, curved, round ■ 

 pointed, posterior protuberance, and the slender forms being often 

 sharp-pointed behind. Protoopalina macrocaudata resembles the 

 stocky form of P. caudata^ while P. orientalis is very similar to the 

 intermediate forms of P. caudata^ which are neither very stocky nor 

 very slender. Protoopalina stevensoni^ from the tropical African 

 Bufo regularise has a posterior process similar to that of P. caudata^ 

 though more spinelike, and the species is probably related to the 

 caudata group. Protoopalina australis^ from an Australian Hyla^ 

 seems also to belong here. 



The occurrence of members of this group in Australia, eastern 

 Asia, Europe and tropical Africa is an indication that the group 



