these may be ta';'en to represent age-groups- V/here 3Jiy given 

 size class is represented by a distinct node in a length-fre- 

 quency distribution, and this mode can be identified in later 

 collections made over an extended period of time, the validity 

 of this association and the identity of the assigned age group 

 is further substantiated. Several investigators have attempted 

 this method of age analysis with the lairvae of other species 

 and have met vd.th var;,''ing degrees of success: their results 

 have sometimes been disputed by other workers „ Gage (1928) 

 using this method Tri.th a "limited series" of larvae of the lake 

 lamprey ( Fetromyzon marinus ) estimated the lens^h of the larval 

 lj_E"e to be not less than four, and probably five years. 



For the purposes of this phase of my study, extensive col- 

 lections of ammoccetes were made in the Ocqueoc River in August,, 

 19k7 (1,331; specimens). All of these collections were m.ade in 

 a restricted area of the river in relatively similar habitats so 

 that growth differences arising from habitats of varying suita- 

 bility would be at a minimum. 



The first collection in August^ 19h7> indicated a discontin- 

 uous size distribution iivherein the range in size of the young-of- 

 the-year (age-group 0, 11.0 to 21.0 mm.) did not overlap with the 

 next size, class. Furthermore, the mode representing the latter 

 grcup (UboO mm.) was very distinct (Figure 55 and Appendix H, 

 Table l). This size class*, termed age-group I^; ranged from 31 to 

 a.bout 59 millimeters. That the size class designated as young- 

 of* the-year was actually composed of age-group was established 

 by follc-ving the groivthj month by months of newly-hatched larvae 

 through their first growing season (Figure 5'6 and Appendix H^ 

 Table 2). The balance of the distribution (59-0 mmo to 151.0 mm,) 

 wa3 continuous with the mode representing age-group I; other modes 

 v?hi.ch might represent older age-groups were obscure. HoTEver^, 

 the oonfig-araticn of the frequency polygon and the total size 

 raTige involved suggested very strongly that at least two additi.onal 

 age-groups (II and III) might be present in the series. 



A second collection, made in October, 19hl 3 gave a length- 

 frequency distribution very similar in character to that obtained 

 for the August series 5 it differed in that the modes and size 

 cilass limit-i. had^ in general^ moved someiAat to the right indi- 

 cating the growth attained in the 60-day interval between ccllec- 

 Mens (f^i.gure 55 and Appendix H, Table l). The Identity of agrt- 

 groups and I was .substantiated by this second series. The 

 balance of the distribution was again continuo'us with the mode 

 representing age- group lo Distinct modes -vvhich might irjlicate 

 additional age groups were lacking except for a pronounced peak' a"^.^ 

 101 millimeters.. However^ considering the range in length that 

 m-5.y be accredited to age-group I (37.0 mm. to 65.0 (?) mm,) in 

 this collection and the size range of the balance of the series 

 (6^»0 (?) mm. to 163.0 ram.) it was still evident that at least two. 

 and probably not more than two^, additional age groups were present 

 i.n the material. 



- ll;9 - 



