tion containing the bacterial DNA — a proxy for 

 the original community. 



One particular gene, or stretch of the DNA, serves 

 as a universal bar code that identifies each of the 

 bacterial strains present in the soil. Fishing out and 

 sequencing the many versions of this bar-code gene 

 and comparing them with similar, yet known, se- 

 quences in gene databases reveals the identities and 

 relative abundances of the various microorganisms. 

 It also shows where the microorganisms from the 

 Tanana River site belong in the bacterial family tree. 



The identity of a microorganism is largely a mat- 

 ter of what it can do, and those functions are, in 

 turn, a matter ot the kinds of proteins the microor- 

 ganism produces. To find out what proteins the 

 Tanana soil bacteria produce, the DNA extracted in 

 the laboratory is converted, in a controlled way, in- 

 to the proteins it codes for, and the proteins are as- 

 sessed for various functions. Typically, the extract- 

 ed DNA is mixed with enzymes that cut it into 

 pieces. The pieces are then inserted into other bac- 

 teria, such as the laboratory workhorse Escherichia 

 coli. Once inside E. coli, they are processed into pro- 

 tein just as if they were part of E. coli's own DNA. 

 Functionally analyzing the new protein is then just 

 a matter of testing the genetically altered E. coli. For 

 example, if the genetically altered E. coli can grow 

 on an antibiotic that would kill ordinary 

 E. coli, the inserted 

 DNA must have 



carried a gene for a protein that somehow disables 

 the antibiotic. 



We also study the Tanana soil microorganisms 

 in a more traditional way: cultured on a Petri 

 plate. When we compare what can be grown using 

 different nutrient sources and growing conditions 

 with what we know is present from the soil DNA, 

 the cultural divide is stark. Bacteria from nineteen 

 different phyla live in the Tanana samples, yet we can 

 grow representatives from only four. 



Microbiologists are sometimes fond of saying, 

 "Everything is everywhere." In other words, bacteria 

 are distributed globally. But as the bacterial census- 

 taking continues, the exceptions, such as our "moun- 

 tain river" microorganisms, keep surfacing. 



The Tanana soils seem to be something of a mi- 

 crobial backwater. We have to look hard to find any 

 Bacillus, one of the most common soil bacteria any- 

 where. Even soil samples from distinct sites on the 

 same Tanana island, or from the same site but just a 

 few inches apart in depth, have different characters — 

 like Scottish villages dominat- 

 ed by one clan or another. 



Although all the sam- 

 pling sites in balsam- 

 poplar forest look sim- 



