Chapter 1 2 



Biolog}; of the Rodents of Enewetak Atoll 



WILLIAM B. JACKSON,* STEPHEN H. VESSEY,* 

 and ROBERT K. BASTIANf 



'Department of Biological Sciences, Bowling Green 

 State Uniuersitt;, Bowling Green. Ohio 43403; 

 fOffice of Water Program Operation, Environmental 

 Protection Agency/. Washington. DC 20460 



INTRODUCTION 



Rodents at Enewetak were casually observed or occa- 

 sionally specifically studied during the nuclear test program 

 (1948 to 1958). However, the rats frequently were 

 misidcntified; no unified analysis was attempted, despite 

 their being the only resident mammals on the test islets. 

 When Jackson was invited to join the University of 

 Washington's resurvey expedition in 1964, the foundation 

 for more than a decade of studies by Bowling Green State 

 University staff and students was established. Ten graduate 

 students participated in these efforts, and data from their 

 theses and dissertations are included in this discussion. 

 The inclusion of one of these students (Temme) in the 

 1978 Northern Marshall Island Radiological Survey permit- 

 ted collection of additional specimens and data from other 

 atolls. 



Origin and Distribution 



Three rodent species are present at Enewetak. The 

 Polynesian rat (Rattus exulans) came with the early 

 Micronesian inhabitants to the atoll. The house mouse 

 fMus musculus) may have arrived with the Japanese 

 administrators before World War II, but major infusions 

 probably came with American activities. The roof rat 

 (R. rattus) apparently arrived with American forces during 

 or after the war. The Norway rat (R. norvegicus), though 

 present elsewhere in the Marsha'ls, has not been observed 

 or trapped at Enewetak (or Bikini). 



Probably the Polynesian rat occurred on most islets 

 used by the Enewetak people for coconut culture, but the 

 combined effects of clearing and construction and the deto- 

 nation of test devices decimated many islet populations of 

 this species. It remains on the less disturbed, more densely 

 vegetated islets. 



The roof rat has flourished on some of the heavily 

 impacted northern islets as well as the main atoll bases 

 (Enewetak, Medren). The survival of this species in Enjebi, 

 within the impact zone of a nuclear detonation (Mike) as 

 well as numerous atomic tests, is hypothesized by Jackson 

 (1969). In an Atomic Energy Commission (AEC) experi- 

 ment it was introduced to Ananij Islet, where it has flour- 

 ished; other introduction attempts were made during the 

 test program but apparently were not successful (Fig. 1). 



The Polynesian rat and roof rat exist allopatrically at 

 Enewetak. Such separation is not total on other Marshall 

 atolls, and one Bikini islet has sympatric populations. The 

 house mouse was found on only three islets (Enewetak, 

 Medren, Japtan) but was in combination with roof or 

 Polynesian rats. 



Cats and dogs existed in varying numbers on islets 

 inhabited by test or administrative personnel (Japtan, 

 Medren, and Enewetak, during our studies). While these 

 animals occasionally caught rodents, they did not seem to 

 have any impact on the populations. Monitor lizards on 

 Japtan caught some rodents, but the impact of such preda- 

 tion was apparently insignificant. Coconut crabs (Birgus 

 latro) were scavengers rather than predators and were 

 observed eating opened coconuts alongside Polynesian rats 

 on Igurin. Reef herons (Egretta sacra) may have been 

 predators, but we did not observe such behavior. 



House Mice 



Mice were found in buildings (occupied or unoccupied) 

 on three islets (Enewetak, Medren, Japtan), and we also 

 caught them regularly in grassland and shrub habitats on 

 these same islets. Detailed studies of house mice were con- 

 ducted on Enewetak and Medren islets (Berry and Jackson 

 1979; Berry et al., 1981). These two populations were 

 genetically different, both in terms of external morphology 

 (pelage) and allozymic variation. Mean heterozygosities (per 

 locus) for these two islet populations were high, 11.4% 

 and 10.9%, respectively. Such variability has been 

 exceeded only in Hawaii (Berry et al., 1981). Selection for 

 certain loci with age, which occurs in nontropical environ- 

 ments, was not observed in these populations. Berry 

 (1979) suggested that such high genetical variability could 



203 



