Elsewhere, the hasty, cut-and-paste strategy caused confusion about the source of information. For example, in a paragraph about University of Georgia scientist Samantha Joye, who found that oxygen levels around undersea oil plumes declined following the spill, Safina copies the following sentence from an Associate Press story: “In an e-mail, Joye calls her findings ‘the most bizarre-looking oxygen profiles I have ever seen anywhere.’” The text implies that Joye sent the e-mail to Safina; readers would have to look up the endnote and then actually read the AP article to know otherwise.

“Yes, I hear what you’re saying about that,” Safina conceded when I pointed out this passage during an interview (he did not have a copy of the book with him), though he defended the clarity of his sourcing overall. “I think it’s pretty clear where people are saying things to me. And where they are not saying things to me, the reference section is the guide to where those things come from,” he said.

The problem is, it’s a rough guide at best. I reviewed the endnotes for the first 150 pages of the 400-page book, and found more than half a dozen instances where references were missing or where Safina had cited blogs or websites that were re-reporting others’ material, rather than the original news stories.

These are just careless oversights, however. The larger concern is the general reliance on other reporters’ work with only an endnote for attribution. Even where articles are cited, the paraphrasing is often so close to the original writing that Safina should be using quotation marks or at least a polite, “according to…” in the book’s text.

For example, Mascarelli pointed out one instance in which Safina listed the health effects of a particular chemical in the dispersant Corexit, writing, “Also it caused breathing difficulties, skin irritation, physical weakness and unsteadiness, sluggishness, convulsions, birth defects, and fewer offspring in mammals.” The Agence France-Presse article that Safina cited listed the effects in the same order, writing, “Other ill effects noted by the [Centers for Disease Control] were breathing difficulties, skin irritation, physical weakness and unsteadiness, sluggishness, and convulsions, as well as birth defects and fewer offspring in mammals.”

A quick visit to the 2-butoxyethanol page at the CDC’s Toxic Substances Portal, reveals these are indeed the symptoms of exposure, but the fashion in which the AFP ordered them is unique. Safina justified copying the list on the grounds of accuracy. “I was trying to be as faithful as possible to the earlier references where I got it, because if you start paraphrasing symptoms, you can make mistakes that give the wrong impression,” he said. “And the same is more or less true with how I used most of those things. I tried to stay pretty faithful to where I got the material and show people exactly where it was from.”

There are cases where the cut-and-paste work cannot be explained by a desire to accurately convey public health information, however. Take another passage Mascarelli found, which reads, “In Gulf Shores, Alabama, thick oil washes up at a state park, coating the white sand with a thick, red stew. ‘This makes me sick,’ says one resident, her legs and feet streaked with crude.” The AP article Safina cites in the endnotes goes like this:

In Gulf Shores, Ala., boardwalks leading to hotels were tattooed with oil from beachgoers’ feet. A slick hundreds of yards long washed ashore at a state park, coating the white sand with a thick, red stew… ‘This makes me sick,’ said Rebecca Thomasson of Knoxville, Tenn., her legs and feet smeared with brown streaks of crude.

“A thick, red stew”? Legs “streaked with crude”? These are not standard medical terms. They are the journalistic stylings and gathered quotes of two AP reporters that Safina unfairly appropriated by forgoing quotation marks or a direct attribution in his text. And this isn’t even the most egregious example. Compare the following passage from Safina’s book and the Los Angeles Times article upon which it is based.

Here’s Safina:

More than a day after the explosion, Stone was finally back on land. “Before we were allowed to leave, we were lined up and made to take a drug tests. It was only then, 28 hours after the explosion, that I was given access to a phone, and was allowed to call my wife and tell her I was OK.”

Then, a few days later, a representative of rig owner Transocean asked him to sign a document “stating I was not injured, in order to get $5,000 for the loss of my personal possessions.” He declined to sign.

Curtis Brainard is the editor of The Observatory, CJR's online critique of science and environment reporting. Follow him on Twitter @cbrainard.