Thanks to Richard Moore for his response. The focus of attention now seems to be shifting to field 378q, about which Richard said "the point I wanted to make was that neither should contain the surname." I'm pretty sure that Richard was intending the case of G. (George) Plumstead and similar ones where only the fuller form of forenames is in question and there is just one surname. If a fuller form is of a compound surname, or includes both forenames and a compound surname, surely all can be recorded? Or most of them. It can get more complex than seems helpful.
As it happens we have locally an update in process to an existing NAR, which is to receive immediate attention (today!). In it a fuller form has several forenames as well as a compound surname recorded in field 378q. The preferred name is "Celia Cruz" but also found is "Ursula Hilaria Celia Caridad Cruz Alfonso" as well as "Úrsula Hilaria Celia de la Caridad Cruz Alfonso" and "Úrsula Hilaria Celia de la Caridad Cruz Alfonso de la Santísima Trinidad." Somewhere I suppose, one has to draw the line and say enough already! The point being that the compound surname "Cruz Alfonso" is recorded in 378q along with as many forenames as good judgment inspires. She did well to use Celia Cruz.
All advice gratefully received! Sincerely - Ian
Ian Fairclough - George Mason University - [log in to unmask]