AMPHIPOD GENUS PARATHEMISTO BOWMAN 383 



Remarks: The straight antenna 1 and the produced inner distal 

 corner of uropod 3 are characteristic of subgenus Parathemislo, but 

 the elongate P5 places P. libellula in subgenus Euthemisto. The 

 species is easily recognized by the pectinate dactyls of P5-7. 



Although Mandt has always been cited as the author of this species, 

 Mandt expressly states that the name and description were prepared 

 by Lichtenstein. The name "libellula" must therefore be attributed 

 to Lichtenstein rather than to Mandt. 



Distribution (figures 18, 19): Circumpolar in the Arctic Ocean. 

 Its distribution in the eastern part of the north Atlantic is given in 

 detail by Stephenscn (1923). He reports it from many localities near 

 Greenland and Iceland. In the Norwegian Sea it is not found below 

 70° N. In the western North Atlantic the southernmost records are 

 from the Gulf of St. Lawrence and the east coast of Newfoundland. 

 In the Pacific Ocean Behning (1939) records it from the Bering Sea 

 and from the Okhotsk Sea as far south as about 46° N. (southeast of 

 Sakhalin Island). Bulychcva (1955) also reports it from the Bering 

 and Okhotsk Seas. I have identified specimens taken in September 

 1958 at four stations in the vicinity of Afognak Island, Alaska, by 

 the University of Washington M/V Brown Bear. No other records 

 from the Gulf of Alaska are known to me. Vinogradov (1956) sum- 

 marizes the known distributional records in his figure 10. 



Aside from P. abyssorum, P. libellula was the only commonly 

 collected amphipod in the 1950, 1951, and 1953 expeditions of the 

 Burton Island. In 1950 it occurred at about one-third and in 1951 at 

 about one-fifth of the stations occupied. The average number per 

 station (100 m. vertical tow) was 0.66 in 1950 and 0.30 in 1951. The 

 average number per positive station was 2.2 in 1950 and 1.5 in 1951. 

 No correlation with variations in temperature or salinity was found. 



P. libellula was taken at 5 of the 25 Burton Island stations in the 

 Prince of Wales Strait in 1953. The average number per station was 

 1.13; the average number per positive station was 1.80. 



Life history: The life history has been studied in some detail 

 by Dunbar (1946, 1957), who points out the importance of P. libellula 

 in the ecology of the eastern Canadian Arctic. It is fed on heavily 

 by the Arctic char and especially by the ringed seal. Dunbar 

 postulates a life cycle of 2 years, with spawning beginning in Sep- 

 tember of the second year and extending through the winter until 

 some time in May or June. The individual is believed to breed 

 once and then to die. If this alternating or 2-phase breeding 

 cycle remained undisturbed, adjacent year classes would be repro- 

 ductively isolated. There is evidence that a few juveniles of each 

 year breed the following winter, so reproductive isolation of the two 

 coexisting year classes is not complete. 



