FISHERY BULLETIN: VOL. 70, NO. 3 



Table 3. — Number of years when given temperature 

 was recorded at the peak of the spawning run. 



Table 4. — Water temperatures (°C) associated with the 

 peak of the shad run and lower and upper temperatures 

 associated with the middle 90% of the shad counted at 

 Holyoke Water Power Company fishway, Connecticut 

 River. 



of the shad appeared at the Bonneville fish lad- 

 ders. In only 3 years did the range exceed 4.5°C 

 (Table 2). 



The only comparable data that we have been 

 able to locate on the Atlantic coast concern the 

 shad passed by the fish lift of the Holyoke Wa- 

 ter Power Company located 138 km from the 

 mouth of the Connecticut River. Daily records 

 of water temperature and the number of fish 

 lifted over the dam show that few shad were 

 passed when the water temperature was below 

 14.0 °C. Peak passage occurred at temperatures 

 ranging from 16.5° to 21.5°C (Table 4). This 

 compares closely with the temperature of peak 



shad passage (16.5°-19.0°C) at Bonneville Dam, 

 233 km from the mouth of the Columbia River. 



Commercial catches provide less direct, and 

 therefore less reliable, data on the timing of 

 shad runs. These can be affected by many ex- 

 traneous factors such as market price, which 

 leads to heavier sampling in the early part of 

 the run, and turbidity of the water, which would 

 have the same effect because catch per effort 

 would be higher early when turbidity is high. 

 Nevertheless, the temperatures found to be asso- 

 ciated with peak commercial catches are in gen- 

 eral agreement with those associated with actual 

 counts at the fishways. 



For the St. Johns River, Fla,, mean weekly 

 shad catches (in pounds) were calculated from 

 daily catch records of the Morris Crab Company 

 for the years 1962 to 1967. These are plotted in 

 Figure 1 with mean monthly river temperatures 

 developed from daily temperature records of the 

 Florida Light and Power Company for the years 

 1960 to 1967. Fishing effort was approximately 

 equal from day to day and year to year from 

 mid-November to mid-March, but after this time 

 market fluctuations resulted in sporadic fishing 

 effort. In order to eliminate error resulting 

 from sporadic fishing effort, catches made after 

 mid-March were omitted from the analysis. 



Shad migrations in the St. Johns River cor- 

 responded closely to the period of lowest annual 

 river temperatures, confirming the earlier ob- 

 servations of McDonald (1884). Few, if any, 

 shad entered the river at temperatures in excess 

 of 20.0°C prior to mid-November, and peak num- 

 bers occurred in mid-January when tempera- 

 tures were at the annual low (15.0°C). As tem- 

 peratures increased, in February and March, the 

 relative abundance of shad declined. 



Massmann and Pacheco (1957) investigated 

 the relationship between temperature and shad 

 catch in the York River, Va., in the years 1953 

 to 1956. From their data we calculated the mean 

 catch of shad per net day, by weekly intervals, 

 February 15 to June 1. The relationship be- 

 tween the timing of shad migrations in the York 

 River and mean monthly river temperatures (de- 

 veloped from daily records for the period 1953 

 to 1962 that were supplied by the Virginia In- 

 stitute of Marine Science) is illustrated in Fig- 



662 



