Ring chromosomes have been described in a number of organisms from viruses to humans, but it is now evident that as a group they are similar only in a superficial, morphological sense. Some are naturally circular, others can be artificially produced while others must be considered sporadically occurring aberrations. The circularity of the lambda bacteriophage chromosome is due to the complementary redundancy of single polynucleotide sequences, which terminate the chromosome, thus permitting it to exist either as a ring or a linear structure. 

The chromosome of Escherichia coli, which is one mm in length and possesses a molecular weight of about two billion, replicates in a circular form although the manner of unwinding of the double helix of DNA during replication still remains a puzzle. The genetic map of such chromosomes would, of course, be similarly circular, for the topography of the map and the physical nature of the chromosome must be consistent with each other. However, such consistency is not found in the T2 or T4 bacteriophage — the map is circular, but the chromosome is a linear structure, 56 µ in length. The T2 chromosomes are circularly and genetically permuted and possess a terminal redundancy amounting to about two percent of the total length. Partial enzymatic digestion followed by annealing produces circular chromosomes, providing proof of the terminal redundancy. 

Therefore, the circularity which occurs naturally in lambda bacteriophage can be artificially induced in the T2 virus. The ring chromosomes of higher organisms, however, must be viewed as aberrant types. They have been studied in Drosophila, maize, and humans, and although they can be perpetuated in certain experimental stocks, they would, in the long run, tend to be eliminated in any situation where they compete with their normal linear homologues. This fact is evident from their behavior, for although they can reproduce in such a manner as to give two un-entangled rings, they also give rise to interlocked rings and to double-sized, di-centric single rings. The fact that cleanly separating rings can be formed at all is, by itself, surprising if it is assumed that the DNA molecule replicates semiconservatively. 

In maize, however, small ring chromosomes freely separate a fair percentage of the time while larger rings have a greater tendency to be interlocked or to form double-sized di-centric rings. Ring chromosomes in humans have involved the X chromosome, some other members of the six-12 group of chromosomes, and chromosomes 17 or 18. 

All are associated with phenotypic abnormalities, and it seems reasonable to suppose that the loss of chromatin accompanying formation of the rings and the irregularity of transmission of these chromosomes are responsible for the observed abnormalities.