males dirring the first half of the riin vras I7.6 inches. This was 

 reduced, as among the females, in the latter half of the season to 

 values r'anging very close to 17.0 inches » This declirie in the 

 sij:e of mig-u-ants may he attributed to an earlier attainment of 

 sexual maturity among larger specimens than among the smaller ones. 



Both range and average length values for the Ocqueoc River 

 run in 19k7 are appreciably lo-mer than those given for all other 

 r'Jns studied. An analysis of the sampling procedure aS undertsJcen 

 in the Ocqueoc River in 'l9hl, hoirever, suggests that it rras 

 selective in. favor of the smaller specimens. During the lfl,tter 

 part of the spaiming season, there was a visible decline in t^he size 

 of males and females on the spawiing grounds. The very smallest 

 specimens seen and collected during 19lil were found on the upper 

 Ocqueoc River spavming grounds during the last vreek in June and the 

 first three days of July. These field observations agree v/ith the 

 data presented for all other runs -whereiji. it vrs-s found that the 

 average size of the upstream migrants decl?jied during the latter 

 lial;? of the upstream movement. ISy collections in the Ocqiieoc River 

 were made prrjnarily at the falls diu:>ing and after the pealc concen- 

 trations of migrants there. Furthermore, collections on the spawn- 

 j_ng grounds v^ere made al^jnost entirely in "blie latter half o;? the 

 period of spawning activityo . For these reasons, I conclude that the 

 data on the ratige and average of size for the Ocqueoc Ri-;?er run in 

 I9I4.7 were influenced by the period duriiig "•.rhich collections vrere made. 



Graphic and ta.bular length-frequency data indicate a platjioirtic 

 distribuT:ion of lengths in the spawning populations entering these 

 streams (i.e., the length-frequency distribution is somewhat flatter 

 thari. a normal curve of distribution Troii3.d be) (Appendix E, Tables 1 

 •and 2; Figure 19). The standard deviations of the lengths of the 

 sexes both separately and combined are high. There is no significant 

 evidence of poly-modality in cirrves constructed for ar^;- of these 

 data. Possible interrpretations of 3uch data may be ambigij.ouSo Platy- 

 kurtic distributions accompanied by high staxidari deviations reflect 

 either hf.gh variability or -some heterogeneity in the sample. In 

 the cr«se of a fish population such as these sea lampreys, this hetero- 

 geneity could be one of either age or origin. By heterogeneity of 

 origiii it is meant -ohat trie lndi->riduals ccmposin.g this: sea lamprey 

 riir., may have come from widely separated localities of diverse char- 

 aot!';.r:Lstics in Lalce Huron. IiTegularities in the length compositxi^n 

 of certain lalce fish taicen mthiji the Lijnits of Saginaw Ba^- a?.one 

 have been attiibuted to thi.? phenomenon accoi-ding to Dr, Ralph Hile 

 (sierbril '^,ommxmi cation) . 



Ttie aet-ei-minstion of -fifcich of the characteristics, heterogeneity 

 o:r b.i^ variability^ is displayed ir the spaTOiing run data 5.3 of 

 oo:csiderable inr[X»rtanee in determining the biology of th3 spocies, 

 Fo'.- example^ it "roul/i be of value to know if tii.j sexuc?,lly ma-'MLre sea 

 j^^ampr-^^rs en-^erijig a stream to opawn are homogenous as to age or 



-56 - 



