AME 24:143-151 (2001)  -  doi:10.3354/ame024143

Sampling-induced artifacts in incubation experiments on biological processing of dimethylsulfide and similarly soluble gases

Richard W. Hill1,*, John W. H. Dacey2

1Department of Zoology, Michigan State University, East Lansing, Michigan 48824, USA
2Department of Biology, Woods Hole Oceanographic Institution, Woods Hole, Massachusetts 02543, USA
*E-mail:

ABSTRACT: Aquatic biological systems that produce or transform dimethylsulfide (DMS), methanethiol, and similarly soluble gases are often studied by means of incubation experiments in closed bottles containing liquid and gas phases. During such experiments, liquid samples are often removed sequentially to obtain time-series of data on microbial numbers, concentrations of nongaseous compounds, or other liquid-phase parameters. The removal of liquid samples from closed bottles affects the concentrations in both liquid and gas phases of 2-phase compounds such as DMS, because sampling alters the liquid:gas volume ratio, inducing chemical redistribution. Three types of problems can arise from multiple sampling in studies of volatile compounds: (1) standardization may become inaccurate because sampling may render experimental and standard bottles incommensurate by creating unequal liquid:gas ratios in the 2 sorts of bottles; (2) changes in concentrations in experimental bottles caused by biological processes may be confounded by sampling-induced changes in concentrations; and (3) concentrations of standards may be altered by sampling from standard bottles. Mathematical expressions of all 3 types of potential distortion caused by sampling are possible, and are useful both for error correction and for designing experiments to minimize artifacts. Mathematical solutions indicate that of the 3 problems, the first is the one that can introduce the quantitatively greatest artifacts. Problem (2) is the most common concern because at least a small confounding of biological and sampling-induced concentration changes is inevitable; the problem is more likely to introduce quantitatively consequential artifacts during falling gas concentrations than rising ones, according to mathematical analysis. The mathematical solutions to Problems (1) and (2) depend on whether the biological cultures are homogeneous. The magnitudes of all 3 problems depend on the partitioning behavior of the gas studied; for instance, in a series of gases that partition principally into the liquid phase, the potential for artifact increases as solubility decreases.


KEY WORDS: Incubation studies · Henry¹s law · Ostwald coefficient · Biogenic gases · Dimethylsulfide


Full text in pdf format