under the conditions described by previous investiga- 

 tors (Ekman 1916, 1927; Hflgboms 1917). The dis- 

 persal over both continents took place later by way 

 of freshwater. It is no more difficult to account for 

 the freshwater dispersal of "marine relicts" than of 

 other widely distributed freshwater forms. If the 

 present species evolved before the termination of the 

 last ice period, the temporary cold glacial lakes 

 may have played an important part in extending 

 their distribution. The present cold-water fauna of 

 Lake Erie would then be the living remnant of the 

 community which populated the fertile lakes in gla- 

 cial times. 



There is evidence of subsequent morphological 

 changes in some of the "marine relict" species now 

 located in isolated areas. This is particularly true 

 in Limnocalanus in which several varieties are now 

 found, the race in Lake Erie being the largest. 



It is probable that the quantitative data on 

 Mysis and Pontoporeia greatly underestimate their 

 true abundance. As in the sea, pelagic net collec- 

 tions do not yield accurate data on such semi- 

 benthic forms as the mysids and grammarids which 

 for the most part spend the greater portion of the 

 year on or adjacent to the bottom. When bottom 

 studies were made in 1928 the largest numbers were 

 always taken in the Helgoland trawls. Six species 

 of larval fishes obtained with the Helgoland trawl 

 were never taken with the meter nets or Petersen 

 trawl in the upper levels. One of these species, 

 Boleosoma nigrum nigrum was found from June 11 

 to August 8 throughout the eastern area in 1928 but 

 did not appear in any of the 1929 collections when 

 the Helgoland trawl was not used. The young of the 

 smallmouth bass Microptertis dolomieu and the scul- 

 pins Cottus cognatus , C. ricei, and Triglopsis thomp - 

 soni were also taken only at the bottom. A similar 

 condition exists at times in the case of Leptodora . 

 Bottom trawls often yielded large numbers in 1928; 

 few or none were taken in net collections made a 

 few feet above bottom. Mud washing disclosed an 

 abundance of individuals living in the mud in shal- 

 low water. As stated in the preliminary report (Fish 

 et a^. 1929) "to obtain a true picture of the macro- 

 plankton fauna of a large lake, it is necessary to tow 

 at various levels with large silk nets, trawl along the 

 bottom with Helgoland or similar apparatus, and to 

 make mud and sand washings along the shore, and 

 perhaps it will be found advisable to examine bot- 

 tom samples from deeper water as well. " 



FISH 



The plankton nets and trawls yielded appreci- 

 able numbers of young fishes which were found in 

 greatest abundance in the littoral zone and over the 

 shallower offshore parts of the lake. The larvae of 

 only the "relict" species favored the cold water of 

 the Deep Hole. The nets yielded 1, 049 specimens 

 representing 18 species in 1928, and 2, 235 specimens 

 representing 14 species in 1929. An additional col- 

 lection of over 20, 000 specimens taken in the west- 

 ern part of the lake was kindly loaned for study by the 

 Ohio Division of Conservation. Of 92 species reported 

 from Lake Erie 62 were identified and described by 

 M. P. Fish (1932). The larval stages of most of these 

 species were previously undescribed. Those not ob- 

 tained are all of extremely rare occurrence, in many 

 cases being represented by single records of capture. 



Young stages and adults of some of the small 

 species of fish are available in large numbers as food 

 for larger predatory forms. Larval and postlarval 

 Notropis atherinoides were taken at every station in 

 the lake often in surprising abundance. Production 

 appears to take place everywhere throughout the sum- 

 mer, but the vast schools of adults were found concen- 

 trated at the surface about the margin near shore. This 

 species was the most common of all fishes found in the 

 stomachs of larger species in the eastern area (Sibley 

 1929). Young of Perca flavescens were also taken in 

 large numbers in the nets and may form an important 

 food item in the littoral zone although only 3 speci- 

 mens were found in the stomach analyses from the 

 watershed and alongshore areas. 



Not one specimen of whitefish ( Coregonus clupea- 

 formis ) or herring ( Leucichthys artedi) appeared among 

 the thousands of larvae and postlarvae in the hauls. 

 Since the time of hatching in these species varies con- 

 siderably in the spring according to temperature, and 

 with observations both in 1928 and 1929 starting soon 

 after the disappearance of the ice in the eastern sector, 

 it was expected that an evaluation might be made of 

 the importance of the spawning grounds on the Canadian 

 side of Long Point as a production center. Yet in a 

 total of 107 hauls made in this sector no larvae were 

 taken in either year. Either the young fish had already 

 passed beyond the stages when they can be captured in 

 meter -net day hauls or the principal nursery for these 

 species in Lake Erie may prove to be in the western 

 part of the lake. 



191 



