236 



FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE 



also estimate the internal radiation for a large fish 

 at 28 mrad./year. Thus, whereas a fish living 

 near the surface (in fresh water the radiation 

 activity from the water itself is estimated at less 

 than 0.5 mrad./year) would receive a total of 63 

 mrad./year, the total dose received would fall 

 rapidly with increasing water depth to 38 mrad. 

 at 10 meters and from 33 to 28.5 mrad./year from 

 20 to 100 meters. A surface-living lake fish would 

 therefore receive about twice the radiation dose 

 of a fish living below 20 meters. 



Most of the salmonids would receive an even 

 heavier radiation dosage than the 63 mrad./year 

 for lake fish at the surface since most of them 

 spend some time in streams, often streams too 

 shallow to afford any shielding effect, in which 

 they would receive additional radiation from the 

 naturally occurring radioactive emitters in the 

 rocks, which varies from about 23 mrad./year for 

 sedimentary rock to about 90 for granite, accord- 

 ing to Folsom and Harley. 



It has been suggested that in part of their range 

 (i.e., in the deep lakes of the Precambrian shield) 

 lake trout might be subjected to considerable radi- 

 ation, particularly in the egg stage or during ex- 

 tended periods spent on the bottom. In the absence 

 of data to refute this suggestion it must be con- 

 sidered as a valid criticism of the above hypothesis. 



To what extent a lowered mutation rate in 

 Cristivomer (which we may perhaps assume from 

 the foregoing discussion of radiation received) 

 could have slowed down the evolutionary processes 

 would be difficult to appraise. An alternate pos- 

 sibility is that Cnstivomei\ during its adaptation 

 to severe conditions in the periods of glaciation 

 that preceded its separation into many isolated 

 colonics, may have lost many of the alleles needed 

 for readaptation to less severe climatic conditions. 

 That this could perhaps be the case is indicated 

 by the ultimate upper lethal temperatures tolerated 

 by various salmonids (Rounsefell, 1958). The 

 young of the other genera all tolerate higher 

 temperatures than the young of Cnstivomer. 



Whether Cristvvoiner or Salvelinus is more an- 

 cient in origin is a moot question that can be 

 argued from different angles. It could be argued 

 that Cristivomer developed from Cristivomer- 

 Salvelinus ancestry in North America while 

 Salvelinus was simultaneously developing in Asia. 

 Later, perhaps, as conditions ameliorated, Salveli- 

 nus invaded North America, either over an Asian- 

 North American land bridge, or from the sea. 



Cristivomer, now isolated in deep lakes, unable 

 without the nest building habit to spawn effectively 

 in streams and unable to tolerate the higher tem- 

 peratures found in most streams, would be unable 

 to make a reciprocal invasion of Asia. 



The theory that Cristivo-mer became recogniz- 

 able in its present form at least as early as the last 

 glacial period is supported by Henshall (1907) 

 writing about the Montana grayling — 



It is very probable that the Arctic grayling was the 

 parent stock from which the Michigan and Montana 

 graylings descended : and from the fact that the habitats 

 of the three species are so widely separated, it is not un- 

 reasonable to suppose that the Michigan and Montana 

 forms were conveyed thence from the Arctic regions dur- 

 ing the Glacial iseriod. This theory is strengthened by the 

 fact that Elk Lake, a half mile from the Montana gray- 

 ling station, is abundantly inhabited by both grayling and 

 the lake trout ( Cristivomer namaycush ) , which latter fish 

 is found nowhere else west of Lake Michigan. 



Sahno might seem to be more ancient in origin 

 than Oncorhi/nchu.s, which is confined to the North 

 Pacific and Arctic Oceans and is much further 

 adapted toward an anadromous existence. Sahno 

 ranges in the western Atlantic from New England 

 to Ungava Bay, thence to southern Greenland and 

 Iceland ; in the eastern Atlantic from Portugal to 

 the White Sea. Since Sahno ( Dymond and Vlady- 

 kov, 1934) is limited on the western side of the 

 Pacific to the Kamchatka Peninsula, it would not 

 seem likely that it had a Pacific origin. Mottley 

 (1934b) suggests that during tlie next to the last 

 glacial period the joint ancestors of Sahno and 

 Oncorhynchus were separated into a Pacific and 

 an Atlantic group, the former evolving into On- 

 corhynchus and the latter into Sahno. During the 

 interglacial ijeriod, Sahno was able to migrate 

 from stream to stream across the continent to the 

 Pacific coast — an impossibility for the strongly 

 anadromous Oncorh ynchus. 



Neave (1958) suggests that Oncorhynchus 

 evolved from Salm-o in the western Pacific, citing 

 in support of his theory the fact that O. masou is 

 more primitive than other species of Oncorhynchus 

 and is more closely related to Sahno. He states — 



In due course the newly evolved offshoot spread back 

 through territories occupied b.v more conservative lines of 

 the anc-estral stock. This process of reinvasion was facil- 

 itated by increased adaptation to ocean life and was ac- 

 companied or followed by a further splitting up into 

 several species. 



None of these explanations suffices to explain 

 fully all of the interrelationships. 



