What are the advantages of raising fry from another female's cluster?

Erich Lorentzen studied this phenomenon many years ago, using Apistogramma borellii females and fry.

He measured the time used for new-born fry to respond on their mother's "gather up" signals.
This time may obviously be a crucial factor to the survival of the fry.

He found that the time decreased when the fry grew older, from 52 seconds for the fry just after they were hatched, to 12 seconds when they reached the age of 6 days.

He then exchanged some of the new-born fry in a cluster with some fry of the same species that were a few days older.

He experienced that the youngest fry tended to follow the older fry (who responded faster) when the mother gave the "gather up"-signals. The result of this was that the entire cluster of fry responded faster to the mother's signals, than a cluster consisting of equal-aged fry.

Result: Stealing fry that are a little bit older than their own fry, and include them in the cluster of own fry, could increase the probability of survival for own fry. And letting another female, with fry that are a little bit older, steal own fry, also increases the survival ratio of own fry (the ones that are included in the other female's cluster).

There may of course be other advantages too, but this seems to be one of them.

Erich Lorenzen (1989): Sind Reaktionen auf einen AAM so fixiert wie seine Schlüsselreize?
Untersuchungen zur Nachfolgereaktionen junger Apistogramma borelli (Teleostei: Percoidei: Cichlidae).
Die Aquarien- und Terrarienzeitschrift, 42, sid 16