716 KELLOGG, SALERNO, AND LATIMER 



incubation of eggs and six static baths for short-duration thermal- 

 tolerance tests. Temperatures in the incubation baths were main- 

 tained within about 1°C by thermostatic mixing valves. Temperature 

 in each bath was monitored hourly with thermistor probes and 

 recorded on a computerized data-aquisition system. Thermal- 

 tolerance test-bath temperatures were maintained within 0.1° C by 

 thermostatic immersion heaters. 



The source of eggs for study varied by species. Atlantic tomcod 

 and alewife adults were collected from the Hudson River and its 

 tributaries and artificially spawned by the dry fertilization method at 

 ambient temperatures. Striped bass eggs were obtained immediately 

 after fertilization from the Texas Instruments hatchery located at 

 nearby Verplanck, N. Y. These eggs were rushed to the laboratory 

 under controlled temperature conditions and aeration. 



The effects of long-term (chronic) temperature elevations on fish 

 eggs were examined by incubating fertilized eggs at ambient and a 

 number of above-ambient temperatures and measuring the hatching 

 success corresponding to each treatment. Eggs fertilized by the dry 

 method were transferred to ambient-temperature baths for a 2- to 

 12-hr period during which it became possible to distinguish fertilized 

 eggs from those not fertilized. Groups of 100 fertilized eggs from 

 each of one to four replicate adult females were placed in 

 screened-bottom incubation containers in the ambient-temperature 

 baths. Five to seven containers from each replicate female were then 

 assigned to each of the eight experimental incubation baths. The 

 percentage of total hatch and percentage of normal hatch (free of 

 visible deformities) were computed for each replicate female at each 

 temperature. The "optimum hatching range" (as defined by 

 Hokanson, McCormick, and Jones, 1973) was determined as the range 

 of temperatures tested over which the percentage of normal hatch 

 was not significantly different from the highest value recorded 

 [P = 0.05, Tukey's multiple range test (Steele and Torrie, I960)] . 

 Also computed was the upper lethal temperature (TL5 q ) determined 

 as the interpolated temperature at which the percentage of normal 

 hatch is 50% of the highest value recorded. 



The effects of short-term (acute) temperature elevations on fish 

 eggs were examined by exposing groups of eggs of several different 

 developmental stages to elevated temperatures for short periods of 

 time and subsequently measuring hatching success. Fertilized eggs 

 were held at ambient temperature before testing. Samples of 50 live 

 eggs were assigned to screened-bottom incubation containers, which 

 were transferred from ambient to several above-ambient tempera- 

 tures for periods of 5 and 30 min. A control sample accompanied 



