hydrographic survey of the Patuxent River, detected the seasonal occurrence 

 in June and July of oxygen -poor water in the deeper areas . This phenomenon 

 occurs during the summer months when density gradients are made more 

 stable by higher temperatures . On the other hand, experimental fishmg with 

 a 16 -foot shrimp trawl yielded consistent catches of hog-chokers ( Achiurus 

 fasciatus) from the test area immediately followmg the explosion, indicating 

 that the water was sufficiently rich in oxygen to sustain fish life. 



Any attempt to predict fish mortality resulting from an underwater 

 explosion must therefore take into account the facts that no fish will be killed 

 unless they happen to be within the lethal range of the explosion and that 

 this lethal range of conventional types of explosives is definitely limited as 

 has already been indicated. Although large kills will sometimes occur, 

 these are most frequent in characteristically "schooling" species, and 

 during those seasons when migrations tend to concentrate the fish in relatively 

 small areas. Practically, a prediction is best based on three factors: (1) 

 recognition of the lethal range of explosives; (2) knowledge of the seasonal 

 abundance of fish and of their habits in the particular locality; and (3) 

 empirical evidence. The third factor provides the soundest foundation. 

 Nevertheless, the second factor is very valuable either in the absence of 

 or in conjunction with empirical evidence. This is well shown by comparing 

 the biological data pertaining to seasonal movements and abundance of fish 

 in area G-U to C-13. That the area (with its own biological characteristics) 

 selected for explosion experiments is quite important in regulating the amount 

 of fish killed is shown in table 7, which indicates the total and average kill 

 (pounds) for one month O^me) in three separate, but not far distant, localities 

 in the same year (1948). These were, the C-ll-C-13 test area. Point 

 Patience, and an area near Barren Island in Chesapeake Bay. The Barren 

 Island data were extracted from a table appearing in an earlier publication 

 by Coker and Mollis (1950). The average kill (119.0' pounds) in each of 

 six explosions near Barren Island far exceeded both the maximum and 

 average kills observed in 19 explosions fired at C-U - C-13 during June in 

 the several years of this study. Likewise, the June average at Patience 

 (38.6 pounds) is well above both the June (1.6 pounds) and total (16.8 pounds) 

 averages at C-U - C-13. The total average kill at Barren Island (21 

 explosions) was 70.7 pounds per explosion, and at Patience (19 explosions), 

 50 . pounds . 



Table 7 



Locality 



Comparison of kill of marketable fish in different areas 

 June, 1948 



Pounds Average 



No. of per 



Explosions Fish Explosion 



C-U - C-13 

 Barren Island 

 Point Patience 



14.0 

 714.0 

 193.0 



1.6 

 119.0 

 38.6 



