1. My later studies confirm my earlier ones as regards the finding of a large curved chromosome in primary spermatocytes, comparable to the so-called sex-chromosome of other forms. 2. The presence of this element is recorded in 963 primary spermatocytes which were sufficiently well prepared to show interpretable detail. It has been observed in many others. It probably exists in all although often obscured by the other chromosomes which tend to stick together. 3. It is, for variously fixed material, surprisingly constant in shape and size, in Langshan, Plymouth Rock and Rhode Island Red fowls. 4. A similarly constant element, differing in form from that of the common fowl, is found in the guinea and in the guinea-chicken hybrid. 5. At the division of the primary spermatocyte this element passes undivided to one pole thus producing two classes of secondary spermatocytes, one with nine and one with eight chromosomes. 6. The element in question is probably a bivalent chromosome formed by the union of two characteristic curved chromosomes which occur in spermatogonial and somatic cells. These elements may be seen to best advantage in the testicular or nephridial cells of 10 to 14 day chick embryos. The remaining chromosomes, typically 16 in number, are usually rod or block shaped. 7. In female chick embryos of 10 to 14 days, a relatively large percentage of dividing cells which were found in the ovarian and nephridial tissues showed a single large curved element. 8. Thus the evidence indicates that the male fowl is homozygous, the female heterozygous for this particular element. 9. The secondary spermatocytes when ready for division display, as a rule, four and five chromosomes respectively. The eight chromosomes which passed to the one secondary spermatocyte have paired to form four, and eight of the nine which passed to the other secondary spermatocyte have paired similarly, leaving the curved one unpaired. 10. The second division is regarded as not a reduction division since in the anaphase the daughter chromosomes often tend each to become bipartite or to resolve completely into two, thus revealing their dual nature. 11. Occasionally the pairing in the secondary spermatocytes is incomplete so that any number between four and nine may appear for division. 12. The odd element after lagging for some time at the equator of the spindle in the secondary spermatocyte, divides. 13. The spermatids which receive four chromosomes frequently pass on to one or more additional divisions. These are regarded as abnormal. Other evidences of degeneration in various spermatids indicate that a considerable number do not develop into normal spermatozoa. It seems probable that only one class of spermatids, that with the odd element, become spermatozoa. 14. The frequency distribution of head-lengths of spermatozoa in four different sets of measurements by two different observers shows no evidence of more than one class of spermatozoa. 15. The cytological evidence as presented in this paper harmonizes with the evidence derived from experimental breeding which shows the female to be heterozygous and the male homozygous for sex and sex-linked characters.