462 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 117 



Barnard 1962d); they are difficult to study because techniques of 

 amphipod study have not benefitted from those of copepodologists 

 or ostracodologists. Often the animals are so small that they are 

 overlooked or are lost in coarse-meshed screens; hence, the best known 

 tropical amphipods are those that are largest, such as species of 

 Elasmopus, Paragrubia, Ampithoe, Maera, and Hyale. Many tropical 

 amphipods autotomize their appendages when preserved, making 

 analysis more difficult. In addition, a high proportion of tropical 

 Amphipoda belongs to genera where gnathopodal characters form the 

 basis of identification and these characters appear only at or after sexual 

 maturity and/or only in males; hence, female and juvenile specimens 

 often cannot be identified without co-occurrence of males, and many 

 samples are therefore worthless. Once life history studies can be 

 conducted it will be possible to identify these nonmales, but the prac- 

 ticing faunistic taxonomist is often unable to solve such problems, 

 especially when collections are so meager. Still another difficulty 

 seems to be the large number of aberrations found in populations of 

 tropical amphipods, especially in such dominant genera as Hyale, 

 Elasmopus, Maera, and Eurystheus. These four genera are highly 

 diverse, the known species in them numbering, respectively, 45, 35, 

 32, and 55 (worldwide). Aberrancies also occur rather more frequently 

 in the species of these genera in colder waters than in other genera as 

 observed by the writer. Because of their high diversity one would 

 expect that the species of these genera would show more variation 

 than species in smaller genera. Such morphological variants probably 

 represent both mutants and ecophenotypes, and they pose problems 

 often insoluble to the morphological taxonomist. 



Potential atoll fauna. — Since atolls have little shelf area, 

 therefore few silts and silty sands in the open sea surrounding them, 

 the atoll fauna is largely epifaunal, composed mainly of species 

 nestling in algae or building tubes on hard objects or inquilinous 

 species commensal with larger organisms. In table 1, I have placed 

 an asterisk on those species in these categories which I believe would 

 occur in Micronesian atoUs. I have eliminated from consideration 

 those species known to burrow in soft bottoms, such as species of 

 Ampelisca, Paraphoxus, Urothoe and all others for which ecological 

 knowledge is poor. 



The ubiquity of the tropical fauna is poorly demonstrated as shown 

 by the occurrence of epifaunal species of amphipods (see table 1): 

 Red Sea only, 11; Red Sea to Indonesia, 45; Indonesia to Hawaii, 63; 

 Red Sea to Hawaii, 31; total, 150. Of those 150 potential atoll 

 species, only 30 are known from the ''Red Sea to Hawaii" (table 1, 

 cols. 1-3, 5-7) or what I call pandemic in the Indo-Pacific. Since 

 the Red Sea may have a number of endemic elements, I have cate- 



