I've been doing some mutant hunting today for some genes of potential interest and because they are fly mutants it's useful to know exactly what the mutation is as they can range from the gene being deleted, a portion being deleted or have a transposable element stuck in them - all of which effects how much of a "mutant" they are. This information is usually in a figure but sometimes, in a busy manuscript, it will be mentioned in the materials and methods. In this case the mutant was referenced as being described in another paper. Fair enough, I thought, and dutifully looked up the other paper. I scour this paper and it merely says "as has been previously described". Not in the paper, so where? In one of their previous papers? In someone else's paper?
This annoys me as I shouldn't have to waste my time chasing up dead-ends. It wouldn't be so bad if this was the first time it's happened but this is a frequent problem when you are trying to find out the exact nature of a mutant, clone construct or methodology. I don't mind them referencing the original paper if it has all the details but people who reference papers that only reference the source material again should hit over the head with a rolled up journal. It's really bad form and strikes of laziness or, worse, hints the author never actually checked the data in the original source.
If editors/reviewers can't be bothered to weed this reference riddles out, then journals should perhaps insist that the information for mutants etc be included as supplementary material - that way they'd have to cite the original if they are printing their data.
I guess the small mercy is that I at least have access to online databases. This kind of nonsense must have resulted in a lot of premature baldness back in the days when people had to go to the library to find papers.