Is there a direct ontogenetic criterion in systematics?