Ethnography is good at some things and structurally poor at others, and knowing which is which is most of what a reference page needs from it.
It reaches the observable well: how the shells are cast, how many land mouth up, who is present, what the sequence of a ceremony is, how a religious house is organised, what an oracle is called. Those are the claims the atlas makes about the sixteen signs, and they are checkable.
It reaches interpretation badly, and not through any failing of method. What a diviner does with a sign depends on the person in front of them and on years of formation, it is not fully articulable even by the practitioner, and much of it is not offered to a researcher in the first place. An ethnographer records what was said; a client receives something else.
So the published record can tell a reader what Ofun, the ten-mouth throw, is within the system, and cannot tell them what it says. That is not a gap this atlas can close by looking harder, and it is not one it should try to close from unsourced material.
Ofun's interpretive content belongs to initiated diviners and to the person receiving a reading, and is not set out here.
There is a further limit particular to interview evidence. A practitioner asked to explain a sign produces an explanation shaped by the question, by who is asking, and by what they judge appropriate to say to that person. The answer is strong evidence about the encounter and much weaker evidence about the practice.