It must be emphasized that population-level information is wholly 

 lacking for most marine organisms potentially impacted by oil spills in 

 the gulf, and that the consequences of physiological, individual-level 

 effects or population-wide effects are assessed almost entirely through 

 the relatively crude measure of relative abundance. That is, the 

 density or abundance of species or individuals of selected species may 

 be measured before (by good chance, occasionally) and after a spill, 

 mortality inferred, and predictive conclusions drawn (e.g. Michael 

 1979). These conclusions have been wrong as often as they have been 

 right. The main reason that predictions have frequently failed is that 

 natural variability in marine populations is enormous and, with or 

 without spills, many species would show extreme seasonal and annual 

 fluctuations in density. Furthermore, sampling methodology is not a 

 small source of variability, and an apparent change in abundance may 

 simply reflect gear or operator differences. 



The reasons for the rather striking lack of correlative information 

 linking spills with biological degradation (changes in abundance, tissue 

 damage, etc.) are multiple. First, and perhaps foremost, is the 

 sampling and natural variability problem. Given the level of natural 

 variability in marine populations and the difficulty of sampling 

 abundance adequately through time, it is extremely difficult to say 

 whether or not a given set of samples indicate changes or trends in a 

 single location, let alone over an entire region. Most marine organisms 

 are patchy in distribution and a local increase or decrease may be 

 insignificant from a community or population perspective. Furthermore, 

 correlative information (e.g. petroleum high here, species "A" low 

 here) is just that: correlative, not proof of any causal relationship. 

 To establish a good correlation is a difficult goal and requires careful 

 experimental design, including proper controls. 



An adequate biological control area in the Gulf of Mexico is 

 difficult to define either spatially or temporally. Hydrocarbons from 

 both spills and from natural sources (Mississippi River, natural seeps) 

 ultimately find their way to the sediments where they are intermingled 

 and changed in nature, or weathered, by various reduction processes. 

 Once in the sediments, the hydrocarbons can be either buried and/or 

 variously resuspended and deposited as a function of bottom turbulence 

 and currents. As part of the sedimentary cycle, the weathered 

 hydrocarbons are distributed over broad areas in low, often uniform, 

 concentrations. Selections of a control area for oil spill assessment 

 studies must take hydrocarbon dispersal and weathering processes into 

 account, both in terms of time and space. 



Contamination of tissues of marine organisms often occurs when 

 hydrocarbon levels are elevated. However, substantial evidence 

 indicates that organisms exposed to high spill levels of hydrocarbons 

 can rid themselves of the pollutants if these levels of exposure 

 decrease (Koons et al. 1977; McAuliffe 1977; Teal 1977; all cited in 

 Sharp 1979). 



69 



