J. Great Lakes Res. 23(4):450-457 Internal. Assoc. Great Lakes Res., 1997 Evidence of Lake Trout Spawning on a Deep Reef in Lake Michigan Using an ROV-Based Egg Collector J. Ellen Marsden l and John Janssen 2 1Illinois Natural History Survey Lake Michigan Biological Station 400 17th St. Zion, Illinois 90066 2Department of Biology Loyola University 6526 N. Sheridan Rd. Chicago, Illinois 60626 ABSTRACT. We report the first evidence of egg deposition by stocked lake trout in deep water in the Great Lakes, on Julian's Reef in Lake Michigan. Historic information, and current data from Lake Supe- rior, suggest that many native lake trout spawned in deep water, and therefore some stocking efforts have been focused on deep reefs. However, no evidence has been previously found to indicate that stocked lake trout spawn in areas deeper than 20 m. The purpose of this study was to find evidence of lake trout spawning on Julian's Reef in Lake Michigan, a deepwater reef (minimum depth 27 m) which was an important spawning site for native strains of lake trout. We used a Phantom S2 ROV equipped with a suc- tion sampler to obtain visual observations of lake trout and to collect eggs and egg predators (slimy sculpins). Five loose lake trout eggs and four sculpins were recovered; the three largest sculpins con- tained nine trout eggs. Adult lake trout were seen at rates above one per minute, in contrast to less than half that rate seen with similar equipment at Burns Harbor breakwall in Lake Michigan. We suggest future modifications of robotic ROV equipment to enhance in situ collections of lake trout eggs and sculpins. INDEX WORDS: Lake trout, Lake Michigan, spawning, ROV, deepwater, reef INTRODUCTION Lake trout (Salvelinus namaycush) populations were extirpated from the lower four Great Lakes by the 1960s due to the effects of overfishing and sea lamprey (Petromyzon marinus). Efforts to restore naturally reproducing populations of lake trout have had limited success except in Lake Superior (Sel- geby 1995). In Lake Michigan, for example, spawn- ing has been noted primarily on shallow, nearshore reefs, including man-made structures such as break- walls; many of these sites were of limited impor- tance for spawning by native stocks, and they have not been the focus of stocking efforts (Holey et al. ICorresponding author. Current address: School of Natural Resources, Aiken Center, University of Vermont, Burlington, VT 05405. Email: emarsden@nature.snr.uvm.edu 450 1995). One of the obstacles to understanding why stocked lake trout fail to reproduce successfully in the Great Lakes is the lack of information about their historic spawning sites. Most information about where native lake trout spawned in the Great Lakes, prior to their extirpation, comes from com- mercial fishermen's records or recollections of con- centrations of ripe individuals (Goodyear et al. 1982, Coberly and Horrall 1980). Such evidence is circumstantial because the nets may have been trap- ping fish in transit to actual spawning grounds. Nevertheless, these historic fishing records are the primary source for predictive information about where spawning by stocked lake trout is likely to occur. Unfortunately, the loss of lake trout from most of the Great Lakes preceded the development of technology capable of studying their spawning habits.