On the possible nature of 85-S ribosomes of bacteria